Item Text

Item text pages: the standard · descriptives · merging with response data · known issues

ImportantAbout these data

IRW item text is reconstructed from published articles, codebooks, and supplementary materials using a largely automated pipeline with partial human review. It is provided for research purposes only. We make no guarantee that the text is accurate, complete, or correctly aligned with the item identifiers in the corresponding IRW response data; the original source document is always authoritative. Users are responsible for verifying item text against that source before drawing substantive conclusions. See Known Issues for documented discrepancies, and please get in touch if you find others.

Note also that the licenses recorded for IRW response data do not extend to the instruments themselves. Inclusion of item text here implies no license to reuse an instrument; copyright and related rights remain with the original rights holders, and users are responsible for obtaining any permissions their intended use requires. Where a table’s wording was released by its source under an open licence (for example CC BY-SA 4.0 or CC BY-NC 4.0), that table’s description on Redivis states the licence, the source and IRW’s changes, and that licence applies to the wording as published here. The licence applied to the item text implies no license to reuse the original source instrument.

Known issues

The below are known issues affecting the match between item text and the IRW tables. 1879 tables are listed. Start typing to search across table names and issue descriptions.

What is listed here. An entry records a concrete mismatch between the item text we ship and the response table: for example, a prompt describing a face-to-face session when the study ran online. Gaps the source itself never published (say, response anchors it never printed) are not listed, and neither are encodings that are correct but surprising, such as an empty item_text on a forced-choice instrument whose wording lives entirely in option_text. Those caveats are on each table’s page under Notes, and all together on Data notes. So a table missing from this list has no known mismatch; that does not mean its source was complete. The one exception is the attribution text some publishers’ terms require (the INEGI entries), which is kept here.

Dataset	Issue
16_personalityfactors	The source’s own instructions describe the scale as “(1) disagree … (5) agree” while its per-item codebook labels it “strongly disagree … strongly agree”; option_text follows the codebook. Item 163 is the one column the codebook omits, so its text comes from the IPIP scale listing rather than the dataset’s own documentation
abdullah_2024_blqol	Responses are stored reverse-coded (7 = ‘never/not related to me’, 1 = ‘always’, higher = less impact of bloating), only the two end anchors are published and only in English, and item QOL4_R (intimate relationships) omits the 133 of 323 respondents the source file codes 0 on that item.
abdullah_2024_bsq_sev24	Item text is the English half of a bilingual instrument table, but the study administered the Malay BSQ-M. The source the study cites for its instrument (Mahd-Abdullah et al. 2021, IJERPH 18:2487, CC BY) lists every item in both languages against the same codes, so the administered Malay wording does exist and is not what ships here; language records the administration as Malay. option_text is empty because the BSQ’s response options are per-item multiple-choice categories that neither paper publishes. SEV241 is marked a reverse-scored item in the source
abdullah_2024_bsq_sevgen	The source papers state per-item scale ranges that do not match the data: items 3/4/6 are documented 1-5 but only reach 4, and item 7 is documented 1-7 but shows 1-4 plus 6-7. SEVG2 is published in the instrument but absent from the table
abdullah_2024_hpbbloat_stress	The source publishes wording only for the final, renumbered 17-item form, which interleaves the five subscales, and the deposited data carry the original SM codes with no labels. Which statement each stress-management code holds (SM1 -> item 4, SM2 -> 9, SM4 -> 13, SM8 -> 17) is therefore inferred by assuming the subscale’s items kept their ascending order; no published source confirms it, so the four texts could be permuted among these codes.
abortion	Four items on when the law should allow abortion, from the 1986 British Social Attitudes Survey, as distributed in the ltm R package. resp is 1 for ‘the law should allow it’. Two caveats from the source: it does not label the 0/1 coding (the Yes/No labels here are inferred from the stem and from which item is least endorsed), and its text for item_1 is an incomplete sentence, shipped unaltered.
abramson_2026_israel_policy	Q3grid_4 was worded differently in the two waves: 2022 respondents read ‘Israel’s actions in the conflict have been taken primarily for defensive reasons’ and 2024 respondents read ‘In general, Israel’s actions in the Israeli-Palestinian conflict have been taken primarily for defensive reasons’. item_text gives both wordings, tagged by wave, and each response row’s wave column says which one that respondent read. Q3grid_3 is reverse-keyed and stored raw.
abramson_2026_mobilization	Although each item reads ‘likelihood of you – …’, the 1-11 response scale is bipolar about direction rather than a likelihood scale: 1-5 run from ‘In support of Israel - 5’ to ‘- 1’, 6 is ‘Will not engage in such action at all’, and 7-11 run from ‘Critical of Israel - 1’ to ‘- 5’, so the midpoint is non-engagement and the ends are opposite stances. The option text is the wave-1 data labels with their spelling ‘Isreal’ corrected; the wave-2 labels mark the same points differently (endpoints plus bare numbers). The item text is the data-file label, not confirmed questionnaire wording.
abukhalaf_2025_disaster_prep	resp is a proportion, not an ordinal level: the source study measured its rising-scale answers and then converted them to percentages, so the 11 values run 0 to 1 in 0.1 steps and mean 0% to 100% likelihood rather than 1st through 11th response option. option_text is empty for every row by design – the source labels no individual point on the slider
abukhalaf_2025_housing_risk	Responses are stored on the study’s 0-1 percentage coding (0, .25, .5, .75, 1); the verbal anchors respondents saw for these possibility and severity questions are not published, so option_text is blank. The source dataset also contains 90 other values (each column’s a single repeated decimal, such as 0.3829, consistent with mean imputation), which the IRW processing script drops, so per-item n is 3-12 below the 816 respondents.
adamczyk_2022_cesd	Respondents answered the Polish CES-D adaptation (Kaniasty, 2003), but neither the deposit nor the paper publishes the Polish wording, so the item text is Radloff’s (1977) English CES-D standing in for what respondents read. The positively worded items CESD4, CESD8, CESD12 and CESD16 are stored already reverse-scored (the depositors’ own depression score is the mean of the stored values), so for those four items the anchors run from ‘most or all of the time’ at 0 to ‘rarely or none of the time’ at 3.
Adherence_Zissette_2018_SDB	These 26 items are a Marlowe-Crowne-derived social desirability scale embedded in a much longer HIV-trial screening questionnaire. The item codes are the screener’s numbering, not the scale’s, which is why they fall in three non-consecutive blocks (q_s_025-032, q_s_056-063, q_s_084-093) rather than a single run. The 1-6 anchors (“Disagree a Lot” to “Agree a Lot”) are the study’s own; its missing-data code is dropped rather than shipped as a response level.
afaya_2020_complications_knowledge	Item wording and the Yes / No / Don’t know anchors are transcribed verbatim from the study’s own deposited codebook (S1 Data), and the item codes are that file’s own column headers, so the code-to-text mapping is tied at the source. Two caveats. The paper’s Methods describes a 23-item Michigan Diabetes Knowledge Test, but the deposited questionnaire is a 66-item battery in seven labelled blocks, of which these seven are the ‘Knowledge of Complications’ block – so the instrument is the study’s own questionnaire rather than the published DKT-23. And although every question has one correct answer, neither the paper nor the codebook says which, so correct_response is left blank rather than asserted by IRW. The questionnaire was administered in English or Dagbani depending on the participant; the English version ships here and the Dagbani form is not published in the deposit.
afaya_2020_diet_knowledge	Item wording and the Yes / No / Don’t know anchors are transcribed verbatim from the study’s own deposited codebook (S1 Data), and the item codes are that file’s own column headers, so the code-to-text mapping is tied at the source; every one of the 39 item-by-level cell counts in the raw spreadsheet matches the live table exactly. Two caveats. The paper’s Methods describes a 23-item Michigan Diabetes Knowledge Test, but the deposited questionnaire is a 66-item battery in seven labelled blocks, of which these thirteen are the ‘Knowledge on Diet’ block, and their wording is the study’s own rather than the published DKT-23 – so the instrument named here is the study’s questionnaire, not DKT-23. And although every question has one correct answer, neither the paper nor the codebook says which, so correct_response is left blank rather than asserted by IRW. The questionnaire was administered in English or Dagbani depending on the participant; the English version ships here and the Dagbani form is not published in the deposit.
afaya_2020_exercise_knowledge	Item wording and the Yes / No / Don’t know anchors are transcribed verbatim from the study’s own deposited codebook (S1 Data), and the item codes are that file’s own data-sheet column headers, so the code-to-text mapping is tied at the source; the source columns’ own response counts reproduce the live per-item means (1.27, 1.14, 2.17, 1.98) and n=330 exactly for all four items. Two caveats. The paper’s Methods describes a 23-item Michigan Diabetes Knowledge Test, but the deposited questionnaire is a 66-item battery in seven labelled blocks, of which these four are the ‘Knowledge of exercise’ block – so the instrument is named as the study’s own questionnaire rather than the published DKT-23. And although the paper says every question has one correct answer, neither the paper nor the codebook says which, so correct_response is left blank rather than asserted by IRW. The questionnaire was administered in English or Dagbani depending on the participant; the English version ships here and the Dagbani form is not published in the deposit.
afaya_2020_footcare_knowledge	Item wording and the Yes / No / Don’t know anchors are transcribed verbatim from the study’s own deposited codebook (S1 Data), and the item codes are that file’s own data-sheet column headers, so the code-to-text mapping is tied at the source. Three caveats. The paper’s Methods describes a 23-item Michigan Diabetes Knowledge Test, and the deposited English questionnaire (S1 File) does contain that multiple-choice test – but it is not what this table holds: these 14 items are the ‘Knowledge of Foot Care’ block of a separate 66-item Yes/No/Don’t-know battery that appears only in the deposit’s codebook and data sheet, so the instrument is named as the study’s own questionnaire rather than as the DKT-23. Two source defects are shipped as recorded rather than corrected: kf11_advice’s text ends mid-word (‘when they have ulce’) and kf10_drytoes carries a double space. And although every question is said to have one correct answer, neither the paper nor the codebook says which, so correct_response is left blank rather than asserted by IRW. The questionnaire was administered in English or Dagbani; the English version ships here, and the deposited Dagbani questionnaire does not include this block.
afaya_2020_general_knowledge	Item wording and the Yes / No / Don’t know anchors are transcribed verbatim from the study’s own deposited codebook (S1 Data), and the item codes are that file’s own data-sheet column headers, so the code-to-text mapping is tied at the source; all 33 item-by-level cell counts computed from the raw spreadsheet match the live table exactly (n=330 per item). Three caveats. The paper’s Methods describes a 23-item Michigan Diabetes Knowledge Test, and the deposited English questionnaire does contain that multiple-choice test – but it is not what this table holds: these eleven items are the ‘Basic general knowledge on diabetes’ block of a separate 66-item Yes/No/Don’t-know battery that appears only in the deposit’s codebook and data sheet, so the instrument is named as the study’s own questionnaire rather than as the DKT-23. Although every question is said to have one correct answer, neither the paper nor the codebook says which, so correct_response is left blank rather than asserted by IRW. And the source’s own typography is shipped unnormalised (inconsistent terminal full stops, mid- sentence capitals in ‘Obesity’ and ‘Increases’). The questionnaire was administered in English or Dagbani depending on the participant; the English version ships here, and the deposited Dagbani questionnaire does not include this block.
afaya_2020_medication_knowledge	Item wording and the Yes / No / Don’t know anchors are transcribed verbatim from the study’s own deposited codebook (S1 Data), and the item codes are that file’s own data-sheet column headers, so the code-to-text mapping is tied at the source; all 30 item-by-level cell counts in the raw spreadsheet match the live table exactly (n=330 per item). Three caveats. These ten items are the ‘Knowledge of medication’ block – medication knowledge, not medication adherence: the same study’s Morisky adherence scale is a separate instrument and is not part of this table. The paper’s Methods describes a 23-item Michigan Diabetes Knowledge Test, and the deposited English questionnaire does contain that multiple-choice test, but it is not what this table holds; these items belong to a separate 66-item Yes/No/Don’t-know battery that appears only in the deposit’s codebook and data sheet, so the instrument is named as the study’s own questionnaire rather than as the DKT-23. And although the paper says every question has one correct answer, neither the paper nor the codebook says which – and two items are self-report questions with no correct answer at all – so correct_response is left blank rather than asserted by IRW. The questionnaire was administered in English or Dagbani; the English version ships here, and the deposited Dagbani questionnaire does not include this block.
afaya_2020_monitoring_knowledge	The item wording for this table comes from the study’s own deposited codebook (S1 Data). The paper’s Methods describes the knowledge measure as the 23-item Michigan Diabetes Knowledge Test, but the deposit actually administers 66 knowledge items in seven blocks, of which this table is the seven-item self blood glucose monitoring block; the shipped instrument name reflects the deposit rather than the paper’s description. The instrument was administered in English or Dagbani and the shipped wording is the English
aguirre_camacho_2021_champion	Item text is the administered Spanish wording, but the 1/5 scale anchors are recorded in English because the source labels only the endpoints and only in English; the middle three points are unlabeled
aguirre_camacho_2021_shai	Item text is the English SHAI-18 original, but the study administered a Spanish translation whose wording the paper never quotes. The instrument has no item stems – each item is a group of four statements – so all text sits in option_text and item_text is empty. The 14/4 split between the two subscales is confirmed against the paper’s subscale statistics, but item order within each subscale rests on HAnx<N> being the instrument’s item N
aiquipa_2026_dgs	The English item text for this Spanish-administered scale is mixed provenance: five items reproduce the deposit appendix’s own English original, while the English for COD3 and COD7 was written by IRW because the Peruvian adaptation changed those two items’ meaning; response- option anchors are blank because the deposit never publishes them
ajaykumar_2023_experience	The item text is the study’s own codebook description of each experience rating (e.g. “Participants’ level of experience with robots”), not the question wording participants read, which was not published. Only the endpoints 1 and 5 of the scale are labelled.
ajaykumar_2023_nasa_tlx	Item text is the official NASA-TLX rating-scale definitions, not the study’s own wording: the study administered a modified NASA-TLX on a 5-point scale and never published what participants saw, so the exact phrasing differs. The paper’s own analyses use three of these six items
ajlan_2025_stemcell_knowledge	The knowledge items are stored as correctness scores rather than the option chosen (2 = correct answer, 1 = not sure, 0 = wrong), so the option text varies by item. Enamel (q23_1) and cementum (q23_3) are stored one point higher by the source file (3 = correct ‘No’, 2 = not sure, 1 = ‘Yes’), so those two items are not on the same 0-2 scale as the other knowledge items.
alan_2018_student_gender_attitudes	The student survey was administered in Turkish; the item wording here is the English rendering from the study’s replication readme, and only the endpoints (1 = strongly agree, 4 = strongly disagree) were labelled.
alan_2018_teacher_extrinsic_motivation	The teacher survey was administered in Turkish; the item wording here is the English rendering printed in the study’s own paper appendix, and item ts_ext_4 is stored reverse-coded (1 = strongly agree) so that higher values mean more reliance on extrinsic motivators for all four items.
alan_2018_teacher_gender_attitudes	Item and response wording is the authors’ own English rendering, published in the paper’s appendix, of a questionnaire administered to teachers in Turkish.
alan_2018_teacher_growth_mindset	The teacher survey was administered in Turkish; the item wording here is the English rendering printed in the study’s own paper appendix, and items ts_gms_1, ts_gms_3 and ts_gms_4 are stored reverse-coded (1 = strongly agree) so that higher values mean a stronger growth mindset for all five items.
alan_2018_teacher_modern_teaching	The teacher survey was administered in Turkish; the item wording here is the English rendering printed in the study’s own paper appendix, and items ts_modern_3, ts_modern_4 and ts_modern_6 are stored reverse-coded (1 = strongly agree) so that higher values mean a more modern teaching approach for all six items; ts_modern_6 responses sit almost entirely at the two extremes (45 at 1, 0 at 2, 2 at 3, 97 at 4).
alcoholhealthwarninglabel_brennan_2022_awareness_harms_followup	resp (0/1) is a derived binary rather than the original response scale: the study collapsed a 7-point agree/disagree item into aware/not-aware, so option_text records that rule rather than a literal option label
alcoholhealthwarninglabel_brennan_2022_intentions_postexposure	Two post-exposure intention items from an alcohol health warning label trial, answered 1 = definitely will not to 4 = definitely will. A third statement in the same block (‘avoid drinking alcohol completely’) is not in this table because the source file holds only a binarised version of it. The wording is quoted from the paper: the deposit itself is behind an openICPSR login.
alexander_2017_dsi	Covers only the Differentiation of Self Inventory’s Emotional Reactivity and Emotional Cutoff subscales (23 of the 43 items). The dsi_1..dsi_23 to original-item mapping is reconstructed from the published appendix order rather than documented in the source, though it matches both the published subscale key and the response data
algner2022_cse	Item text is the English CSES original; the study administered a German translation whose exact wording could not be located
algner2022_oss	The study administered a German version of the Oslo Social Support Scale short form, but the item text here comes from the English original, not the German respondents read. Only four of the six items have text: item 2 is from Algner’s paper, items 5 and 6 are the English original as quoted by a Chinese validation study, and item 1 is a lower-confidence back-translation from a Brazilian adaptation. Items 3 and 4 are blank because no source gives their literal wording.
ali_2021_isi	The item text is the canonical Insomnia Severity Index wording (Morin et al., 2011), not the wording in the study’s own S1 File. The extraction silently normalised the source’s column headers toward the canonical form, so six of the seven shipped items differ from the deposit: the source reads “Difficult falling asleep”, “Difficult staying asleep” and “Problems waking up too early” where the shipped text reads “Difficulty falling asleep”, “Difficulty staying asleep” and “Problem waking up too early”. Those headers are abbreviated spreadsheet labels rather than administered sentences, so neither form is certainly what respondents read, and the instructions are the standard ISI preamble because the study publishes none. Items 1-3 carry fewer responses (103/94/105 against 167-168) because those severity anchors are sometimes skipped as inapplicable.
alkouri_2025_icu_stressors	The source questionnaire changes the wording of its second response option partway through the scale – items 01-14 offer “Barely” and items 15-29 offer “Rarely” for the same value 2 – so option text varies by item; transcribed as the source gives it
almuqbil_2022_epds	Participants completed the Arabic EPDS translation (Ghubash et al. 1997); the item text here is the English wording used to label the items in the study’s own data file, so it is not what respondents read
alqerem_2024_diabetic_health_literacy	Seven items from the Jordanian Diabetic Health Literacy Questionnaire, on understanding diabetes information and communicating about diabetes; the wording is the study’s own English labels, while patients answered an Arabic version that was not published. THE RESPONSE CODES ARE NOT A CLEAN ORDERED SCALE: the paper analysed a four-point scale, but the deposited data use five codes, and for the four information items the published frequencies are only reproduced if code 2 is the paper’s third level, code 3 its fourth, and codes 4 and 5 together its second, so higher resp does not mean higher literacy and no anchor labels are shown.
alsecypiamh_wu_2022_cps	The item text here is the administered Chinese, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors.
ALSECYPIAMH_WU_2022_PHQ	The item text, the ‘Over the last 2 weeks’ stem and the 0-3 ‘Not at all’ to ‘Nearly every day’ options are the standard published English PHQ-2 wording, not the study’s own materials. The study administered the Chinese PHQ-2-C (Liao et al. 2017), so this is not what respondents read, and language records the administration as Chinese. The study’s raw data file carries no item labels and its paper is paywalled, so the administered wording could not be recovered from the study itself.
alsuhibani_2022_gcbs_extra_s2	Only GCBS16 (‘Princess Diana had to be killed…’) is one of the two extra items Study 2 added to the GCBS. GCBS17 is the standard GCBS item ‘A lot of important information is deliberately concealed from the public out of self-interest’ (item 15 of the 15-item scale). The other extra Princess Diana item (‘Princess Diana faked her own death…’) is GCBS7 in the Study 2 source file and is not in this table.
alsuhibani_2022_loc	The paper’s Methods describes this as a five-point scale, but the live data and the study’s own SPSS file both show a genuine 6-point “Strongly disagree” … “Strongly agree” scale; the data file was followed rather than the paper text
alsuhibani_2022_paranoia_s3	The study’s data file labels the five response options Strongly disagree to Strongly agree, which is what is shipped, but the paper states this scale used the same response format as its conspiracy scale (‘Definitely not true’ to ‘Definitely true’), so respondents may have seen the latter wording.
alves_2017_hamd17	Item text follows a canonical reproduction of Hamilton’s original (1960) HDRS-17, which scores item 9 (Agitation) 0-2, but this study’s data scores it 0-4 – so HamD9Baixa responses of 3 and 4 have no option label. Item text is also a canonical reproduction rather than the study’s own materials, and item-to-text alignment follows the instrument’s standard order rather than an explicit map
amarilla_2020_lawton_brody	resp (0/1) is the Lawton-Brody dependency score rather than the administered response: each domain offers 3-5 instrument levels that collapse to able/unable, so option_text names the scored category and lists the levels sharing it instead of a single literal option label. The study’s data supplement also records only the total IADL score, so the item-to-domain mapping follows the standard Lawton & Brody (1969) ordering rather than an explicit map, and item text comes from the canonical instrument
amatus_cipora_2024_fsmas_se	Items SE1-SE4 are stored reverse-keyed in the source, so on every item a higher resp is the less stereotyped answer. The codebook’s note that a higher sum means more stereotype endorsement does not hold for the stored values
american_multiracial_face	Each item is one photograph rated on one attribute, and the item codes are arbitrary integers assigned when the study’s 19 separate rating files were stacked – they carry no meaning and must not be read as an item order. item_text is therefore a constructed description of the stimulus rather than transcribed text; the verbatim rating question is the section_prompt. The AMFD PhotoID named in item_text is not stated in the source materials and was recovered by matching per-face rating means against the published norming table, and 9 faces (171 items) were never released in the AMFD and so carry no PhotoID. Only scale points 1 and 7 were labelled in the original survey, so option_text is blank for 2-6. Separately, the table has no Black-prototypicality ratings for the smiling photographs: the source repository provides them, but the IRW processing script does not read that file
an_2020_efl_self_regulated	The table name signals one instrument but the table pools three administered in the same survey: the TSELSS strategy items (Q1-Q26), an English-learning enjoyment/environment block (Q27-Q33), and a 16-item English language self-efficacy questionnaire (Q34-Q49). Item text is the original Chinese from the study’s own data file. Only the two endpoints of the 7-point scale are labelled in that file, so option_text is blank for values 2-6
anh_2026_ai_adoption	The questionnaire was administered in Vietnamese (translated and back-translated), but no Vietnamese wording is published in the paper or its only supplement, a numeric data CSV, so the item text here is the English source wording the paper prints in its Table 3 rather than the sentences respondents read. Table 3 is published only as an image, so the English was transcribed by eye and may carry transcription slips in punctuation. Only scale points 1 (strongly disagree) and 5 (strongly agree) carry labels; points 2-4 are blank.
anh_2026_digitaltrust	The questionnaire was administered in Vietnamese (translated and back-translated), but no Vietnamese wording is published in the paper or its only supplement, a numeric data CSV, so the item text here is the English source wording the paper prints in its Table 3 rather than the sentences respondents read. Table 3 is published only as an image, so the English was transcribed by eye. Only scale points 1 (strongly disagree) and 5 (strongly agree) carry labels; points 2-4 are blank.
anh_2026_finliteracy	The questionnaire was administered in Vietnamese (translated and back-translated), but no Vietnamese wording is published in the paper or its only supplement, a numeric data CSV, so the item text here is the English source wording the paper prints in its Table 3 rather than the sentences respondents read. Table 3 is published only as an image, so the English was transcribed by eye. Only scale points 1 (strongly disagree) and 5 (strongly agree) carry labels; points 2-4 are blank.
anh_2026_finsocialization	The questionnaire was administered in Vietnamese (translated and back-translated), but no Vietnamese wording is published in the paper or its only supplement, a numeric data CSV, so the item text here is the English source wording the paper prints in its Table 3 rather than the sentences respondents read. Table 3 is published only as an image, so the English was transcribed by eye; the source’s own sentence-medial periods (e.g. FS1 ‘growing up at home. my family’) are kept as printed. Only scale points 1 (strongly disagree) and 5 (strongly agree) carry labels; points 2-4 are blank.
anh_2026_finwellbeing	The questionnaire was administered in Vietnamese (translated and back-translated), but no Vietnamese wording is published in the paper or its only supplement, a numeric data CSV, so the item text here is the English source wording the paper prints in its Table 3 rather than the sentences respondents read. Table 3 is published only as an image, so the English was transcribed by eye; the source’s own punctuation errors (e.g. FW4 ‘At the end of the month. I have extra money.’) are kept as printed. Only scale points 1 (strongly disagree) and 5 (strongly agree) carry labels; points 2-4 are blank.
anjum_2022_gad7	The IRW table covers 6 of the GAD-7’s 7 items: item 1 (“Feeling nervous, anxious, or on edge”) is present in the study’s data file but was not included in the IRW release. Two item wordings and the response anchors follow that file’s own labels, which differ from canonical GAD-7 wording – “not being able to sleep or control worrying” for “stop or control”, “hart to sit still” for “hard”, and “More than half of the days” for “More than half the days”. The instructions are the canonical GAD-7 stem; the data file carries none
anxiety_gastro_symptoms	Fifteen statements about anxiety and hypervigilance regarding gut symptoms, rated 1 = Strongly Disagree to 5 = Strongly Agree. Items aehas1-aehas9 form the anxiety subscale and aehas10-aehas15 the hypervigilance subscale. The source file codes the same anchors 0-4; the IRW values are one higher.
AOMT_BR_SF_EDPANAB_Geiger_2021_AOT	The table carries 14 item codes for a 10-item scale. AOT3_rev, AOT5_rev, AOT7_rev and AOT8_rev are reverse-coded duplicates of AOT3, AOT5, AOT7 and AOT8 rather than additional items – the same statement, with the response options in reverse order – so they carry their parent item’s text and reversed option_text
AOMT_BR_SF_EDPANAB_Geiger_2021_BRS	The table name signals the Bullshit Receptivity Scale alone, but it combines that scale’s 10 items (BSR*) with the 10 motivational-quotation control statements (MOT*) from the same study; both blocks were rated on the same “profoundness” scale
AOMT_BR_SF_EDPANAB_Geiger_2021_RF	The shipped wording is the English version read by the 259 North American respondents; the 248 Bulgarian respondents answered a validated Bulgarian translation, and in the North American survey the disease problems (RCG4, RCL2) named the respondent’s own country, shown here as ‘Canada/the United States’.
APFCompact_Ptacek_2024_DASS-21	The item text is the canonical English DASS-21 wording, but the study administered a Czech translation that is not published in its OSF deposit or paper, so these are not the sentences respondents read. The item codes are the deposit’s own column headers, which carry the canonical DASS-21 item number and subscale letter.
arabaci_2025_skill_diversity	Respondents answered a Turkish version of the scale, which is not published with the study; the item text is the study’s own English appendix. The appendix lists the four statements without numbers, so which statement belongs to which item code follows the listing order, and the paper publishes no item-level statistics that could check it. Only the endpoints of the 1-5 scale are labelled.
arabaci_2025_turnover_intention	Respondents answered a Turkish adaptation (Örücü and Özafşarlıoğlu, 2013) whose wording is not published with the study; the item text is the study’s own English appendix, and the paper quotes the items in slightly different English elsewhere (‘I often think about quitting my current job’). The appendix lists the statements without numbers, so codes follow the listing order; the data confirm which statement is item 1 but cannot separate items 2 and 3. Only the endpoints of the 1-5 scale are labelled.
argentina_2012_aging_ageism	The English statement and response-option wording for this table was produced by IRW, not by INDEC, which publishes the ENCaViAM 2012 in Spanish only; the administered Spanish is in the base fields. The Spanish is taken from INDEC’s Diccionario de variables, not from the fielded questionnaire, which the documentation says is appended but is not; no interviewer lead-in was recovered. The resp direction differs by item: the IRW processing script reverses re01, re02, re04 and re05 so that 1 = No and 2 = Sí, and leaves re03, which is worded in the respectful direction, as coded (1 = Sí, 2 = No). ‘Ns/Nc’ answers are set to missing, and the 17 respondents who gave it on all five statements are absent from the table.
argentina_2012_aging_change	The English item and response-option wording for this table was produced by IRW, not by INDEC, which publishes the ENCaViAM 2012 documentation in Spanish only; the administered Spanish is in the base fields. The Spanish is taken from INDEC’s Diccionario de variables, not from the fielded questionnaire, which the documentation says is appended but is not; the stem ends in an ellipsis (‘En relación al año pasado, ¿usted diría que su salud…’) and is completed by the option read aloud (‘…ha mejorado?’). The IRW processing script reverses the source codes, so resp 3 = has improved, 2 = the same, 1 = has worsened.
argentina_2012_aging_dependence	The English item and response-option wording for this table was produced by IRW, not by INDEC, which publishes the ENCaViAM 2012 documentation in Spanish only; the administered Spanish is in the base fields. The Spanish is taken from INDEC’s Diccionario de variables, not from the fielded questionnaire, which the documentation says is appended but is not, so no interviewer lead-in was recovered. The IRW processing script reverses the source codes (1 Sí, 2 No), so resp 2 = Sí (needs help) and 1 = No.
argentina_2012_aging_health	The English item and response-option wording for this table was produced by IRW, not by INDEC, which publishes the ENCaViAM 2012 documentation in Spanish only; the administered Spanish is in the base fields, with ‘regular’ rendered ‘fair’ and ‘mala’ ‘poor’. The Spanish is taken from INDEC’s Diccionario de variables, not from the fielded questionnaire, which the documentation says is appended but is not; each stem ends in an ellipsis and is completed by the option read aloud (‘…excelente?’). The IRW processing script reverses the source codes, so resp 5 = excellent, 4 = very good, 3 = good, 2 = fair, 1 = poor.
argentina_2012_aging_instrumental	The English item and response-option wording for this table was produced by IRW, not by INDEC, which publishes the ENCaViAM 2012 documentation in Spanish only; the administered Spanish is in the base fields, and ‘remis’, an Argentine hired-car service, is kept in the English with a gloss. The Spanish is taken from INDEC’s Diccionario de variables, not from the fielded questionnaire, which the documentation says is appended but is not, so no interviewer lead-in was recovered; dep03_05 to dep03_07 open with ‘Por problemas de salud’ and dep03_01 to dep03_04 do not. The IRW processing script reverses the source codes (1 Sí, 2 No), so resp 2 = Sí (needs help) and 1 = No.
argentina_2012_tobacco_harm	The English item and response-option wording for this table was produced by IRW, not by INDEC, which publishes the Encuesta Mundial de Tabaquismo en Adultos (EMTA) 2012 in Spanish only; the Spanish in the base fields is the question wording printed in INDEC’s data codebook, not a fielded questionnaire, so interviewer framing is not included. The IRW processing script reverses INDEC’s coding (INDEC 1 ‘Sí’, 2 ‘No’), so here resp 2 = Sí and 1 = No, and it drops ‘No sabe’ (don’t know) to missing, which removes 43-57% of respondents on items h02d, h02e, h02g and h03.
argentina_2012_tobacco_policy	The English item and response-option wording for this table was produced by IRW, not by INDEC, which publishes the Encuesta Mundial de Tabaquismo en Adultos (EMTA) 2012 documentation in Spanish only; the Spanish in the base fields is the question wording printed in INDEC’s codebook, as the fielded questionnaire with its interviewer framing was not recovered. Note also that resp is stored reversed from the source codes (1 = En contra, 2 = A favor, where INDEC codes 1 = A favor, 2 = En contra), and that ‘No sabe’ and ‘Se niega’ responses were dropped to missing by the IRW processing script.
argentina_2013_reproductive_awareness	The English item, instruction and response-option wording for this table was produced by IRW, not by INDEC, which publishes the Encuesta Nacional sobre Salud Sexual y Reproductiva (ENSSyR) 2013 in Spanish only; the administered Spanish is in the base fields. Note also that resp is stored reversed from the questionnaire’s codes (1 = No, 2 = Si, where INDEC codes 1 = Si, 2 = No), and that ‘Ns/nc’ (don’t know / no answer) responses were dropped to missing by the IRW processing script.
argentina_2013_reproductive_transmission	The English item and response-option wording for this table was produced by IRW, not by INDEC, which publishes the Encuesta Nacional sobre Salud Sexual y Reproductiva (ENSSyR) 2013 in Spanish only; the administered Spanish is in the base fields. Note also that resp is stored reversed from the questionnaire’s codes (1 = No, 2 = Sí, where INDEC codes 1 = Sí, 2 = No), and that ‘Ns/Nc’ (don’t know / no answer) responses were dropped to missing by the IRW processing script.
argentina_2017_victimization_confidence	The English institution names, question and response-option wording for this table were produced by IRW, not by INDEC, which publishes the ENV 2017 in Spanish only; resp is stored reversed from the source codes (4 = Muy confiable, 1 = Nada confiable), and the question offers ‘algo confiables’ while INDEC labels response 2 ‘Confiable’.
argentina_2017_victimization_disorder	The English question and response-option wording for this table were produced by IRW, not by INDEC, which publishes the ENV 2017 in Spanish only; resp is stored reversed from the source codes (2 = Si, 1 = No).
argentina_2017_victimization_fairness	The English question and response-option wording for this table were produced by IRW, not by INDEC, which publishes the ENV 2017 in Spanish only; resp is stored reversed from the source codes (4 = Muy de acuerdo, 1 = Muy en desacuerdo).
argentina_2017_victimization_fear	The English question and response-option wording for this table were produced by IRW, not by INDEC, which publishes the ENV 2017 in Spanish only; No aplica responses (e.g. 18011 of 29303 for the educational-institution item) were dropped from the table.
argentina_2017_victimization_performance	The English question and response-option wording for this table were produced by IRW, not by INDEC, which publishes the ENV 2017 in Spanish only; resp is stored reversed from the source codes (4 = muy buena/muy bueno, 1 = muy mala/muy malo).
argentina_2017_victimization_precaution	The English question wording for this table was produced by IRW, not by INDEC, which publishes the ENV 2017 in Spanish only; resp is stored reversed from the source codes (2 = Si, 1 = No).
argentina_2017_victimization_trend	The English question wording for this table was produced by IRW, not by INDEC, which publishes the ENV 2017 in Spanish only; resp is stored reversed from the source codes (3 = Aumento/increased, 2 = Se mantiene igual/stayed the same, 1 = Disminuyo/decreased).
armour_2017_pcl5	The item wording is the published PCL-5 standard form from the National Center for PTSD, not the study’s own survey screens, which were not published; the paper confirms past-month ratings about the respondent’s worst event on the same five 0-4 anchors.
arnulf_2022_general_knowledge	The general-knowledge test was open-answer, so participants did not choose from the options shown here: option_text gives the Correct/Incorrect scoring labels the study’s data file attaches to 0 and 1. Item text is verbatim from that file, including its “principle street for finance in New York” where the paper’s prose reads “principal”
arora2025_blueq_synchronous	The study’s data file records only numeric codes, so the response anchors and the section prompt are taken from the published BLUE-Q instrument. That instrument’s prompt describes a “face-to-face component”, while this study administered the synchronous sessions as online workshops – the item text itself, which is verbatim from the study’s own spreadsheet, refers to “face-to-face online workshops”
art	The Author Recognition Test presents a checklist of names with no printed response labels, so option_text (“Checked”/“Not checked”) describes the check-box rather than quoting the instrument, and item_text repeats the author name shown on the checklist. Only the 50 real-author items are in the IRW table; the 50 foils were dropped upstream. The name “T.S. Elliot” is kept as the source data spells it (canonically “T. S. Eliot”)
artistic_preferences	The table name describes only the first of three instruments administered in the same session: items 1-30 are the Artistic Preferences Scales, items 31-40 the Ten Item Personality Inventory, and items 41-56 a 16-word vocabulary check list. Items 1-30 have no text stem at all – each is a pair of paintings, shown as images, that the respondent chooses between and rates the strength of the preference; the images are distributed with the source dataset. Only the two extreme points of that 1-5 scale carry anchors, so option_text is blank for 2-4
arzamoncunill_2023_epq_admin	Item text for the 14 items kept in the final questionnaire is the administered Spanish with English alongside (12 of the 14 English renderings are IRW’s own translations); the 7 items dropped during the study’s factor analysis never had their wording published, so E5, E6, E38, E45 and E46 carry only the study’s short English category label, and E37 and E39 have no item text because the study’s supplement and its paper contradict each other about which of the two is ‘Access from different devices’ and which is ‘Different data hosting options’.
arzamoncunill_2023_epq_clinical	Ten of the 22 items here are not the wording respondents read. The study ran a 43-item pretest pool but published full item wording, in Spanish, only for the 26 items kept in its final questionnaire, while this table holds all 22 clinical-care items of the pretest pool. The 12 that survived carry their administered Spanish from Appendix S3; the other 10 have no published wording in any language, so they keep the short category descriptor the study’s supplement gives for each item number (“Editable agenda”, “Videoconference”), which names what the item is about rather than quoting it. Those 10 are identifiable by an empty item_text_translated. Two of the 12 English translations are the authors’ own, printed in both languages in Appendix S1; the other 10 were produced by IRW rather than by the study. The 7-point anchors are published in English only (“very much disagree”, “very much agree”), so option_text is English throughout. The study’s own data file carries no variable or value labels; the item numbers are matched exactly, and the 12 Spanish items are tied to their codes through the supplement’s number-to-descriptor map, cross-checked against the retained-item list in the study’s factor analyses
aslec_insomnia_wang2025	The item text here is the administered Chinese, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. instructions and option_text are already English in the published table, so only the Chinese item text carries a translation.
assanangkornchai_2022_cannabis_benefits	The English translation is the study’s own and is looser than the Thai administered wording in places (e.g. item V514 ‘Reduce nausea and vomiting’ omits ‘from chemotherapy’); the deposit codebook’s label for V503 (‘spasticity in multiple sclerosis’) is wrong – V503 is ‘Improve spasticity’ and the multiple-sclerosis spasticity item is V516.
assanangkornchai_2022_cannabis_harms	The English translation is the study’s own and is terser than the Thai administered wording in places (e.g. V526 omits ‘increases the risk of accidents’).
audit_BrummerHoffman_2021	AUDIT-C administered in Brazilian Portuguese; the English alongside is the study’s own translation. Two points to know before using the scores. The third item (six or more drinks on one occasion) is scored in the standard AUDIT-C direction, 0 = Nunca to 4 = Todos ou quase todos os dias – the deposit’s English questionnaire prints those options in the reverse order, which does not match the data. And on the second item, ‘Nao bebo’ (I do not drink) and ‘1 ou 2’ were separate options on the form that both score 0, so a 0 there does not distinguish abstainers from light drinkers; the two are shown as one label because the value cannot tell them apart. The AUDIT is a World Health Organization instrument (WHO/MSD/MSB/01.6a, Babor et al. 2001) and WHO reserves all rights in it; WHO permits free reproduction for non-commercial purposes, which is the basis on which it appears here. The Brazilian Portuguese wording is the study’s own translation, published in its OSF deposit, not a WHO translation.
audretsch_2021_entrepreneurial_ecosystems	Item text is the untruncated form of the study data file’s variable labels, recovered from its Stata do-file because the .dta labels are cut off at Stata’s 80-character cap. For the ee_quality item the paper’s own section 4.2 quotes a different wording (“There is a strong focus on growth-oriented and productive entrepreneurship activity in my region (city)”) from the data-file label that is shipped here (“There is a strong entrepreneurship ecosystem (EE) in my region (city)”). Several labels are ungrammatical or unclosed in the source and are transcribed as they stand. Only the endpoints of the 1-7 scale (1 = very weak, 7 = very strong) are documented, so the intermediate options are blank
autonomysupport_mokken	The item “Decide” is reverse-worded in the original questionnaire, but the responses in this table are the raw, unreversed scores, so the same anchors apply to it as to every other item. Item text is the English rendering the mokken package documents for a questionnaire administered in Dutch. Only the two extreme anchors are documented (1 = not at all/never, 5 = certainly/always)
avilesgonzalez2019_ces	Item text comes from the study’s SPSS file, which stores the English Caring Efficacy Scale wording even though the respondents were Italian nurses, so it is not what they read. Several typographical errors in those labels are preserved here as the source has them – one item is missing its opening “If”, another reads “Ioften”, and several begin with a lowercase “i”
baaziz_2023_sms2	The text shown for item 16 is probably not the item respondents answered, and should not be used as the wording of that item. The paper prints “So that others praise me for what I do” at position 16, which is external-regulation content, but in the data item 16 behaves as amotivation: it correlates 0.88-0.90 with items 17 and 18, the other two amotivation items, and essentially zero (0.02, 0.04, -0.03) with the three external-regulation items 13-15. Position 16 in the SMS-II is an amotivation item, so the paper’s table appears to misprint its wording. The misprint is in both languages – the Arabic printed at position 16 carries the same external-regulation content – and both are shipped unchanged rather than substituted with a guess. item_text is the administered Arabic from the paper’s Table 7, with the authors’ own English back-translation in item_text_translated. The 7-point scale anchors are an exception: the paper gives them only in English (“Not at all true”, “Very true”), so option_text holds English rather than what respondents read, and its _translated twin is empty. Only the endpoints are documented, so options 2-6 are blank
baka2023_jcs	Item stems are in Polish as administered, but the response anchors carried by the study’s data file are English glosses rather than the Polish wording participants saw
baka2023_olbi	Only 8 of the published OLBI’s 16 items are present in the study’s data file, and those 8 are what this table covers. Item stems are in Polish as administered, but the response anchors carried by the data file are English glosses rather than the Polish wording participants saw
baka2023_uwes	Item stems are in Polish as administered, but the response anchors carried by the study’s data file are English glosses rather than the Polish wording participants saw
bakker_2020_rses	The item codes for this table must not be read as item content. Seven of the ten columns in the study’s SPSS file carry a variable label that contradicts the column name, and the data show the labels are the reliable half: the column named goodqualities303 is labelled “All in all, I am inclined to feel that I am a failure” and averages 1.47 on a 1-4 Strongly Disagree..Strongly Agree scale, while nogood302 is labelled “I feel that I have a number of good qualities” and averages 3.27. Every negatively-worded label sits at the floor and every positively-worded one at the ceiling, which a positively-worded self-esteem item could not do in a general sample. Read in column order the labels reproduce a standard circulated administration order of the scale, so the names appear to have been assigned by numbering the columns in Rosenberg’s original order without noticing the form was permuted. The text shipped here follows the labels; join on the item code and read the text, not the code. Responses are stored raw and unreversed. Note also that the paper’s Methods states the scale ran 1 = strongly agree to 4 = strongly disagree, the reverse of the data file’s own value labels; the data follow the file
bakumenko_2023_adyghe_values	The item text here is the administered Russian, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. Only one option label exists in the published table for a 1-5 scale, so the other four points carry no option text.
balmas2018_leader_attitudes	Only the poles of each 1-5 scale are labelled, following the study’s codebook, so option_text is present at 1 and 5 and blank at 2-4. Bracketed alternatives such as [US; Israel] and [America's; Israel's] are the study’s own notation for its two samples and ship unresolved, so the item text is not what either group saw verbatim.
balmas2018_leader_personality	Eleven of these items are bipolar adjective pairs rated 1-5 (for example “Untrustworthy-Trustworthy”), and the study’s codebook labels only the two poles, so option_text is present at 1 and 5 and blank at 2-4 by design. Bracketed alternatives in the text – [Germany; Israel], [US; Israel] – are the study’s own notation, not an IRW artifact: the same item went to a US sample and an Israeli sample with the country substituted, and it ships unresolved, so the wording here is not what either group saw verbatim. All items are oriented positive-high as published and none were reverse coded; the codebook’s don’t-know/no-answer code is dropped, matching the 1-5 range shipped.
bang_2023_self_esteem	The item order in this table was reconstructed, not read from any key, and the item text is not what respondents saw. Neither the study’s data file (bare codes S1-S10) nor the paper ties any code to an item, so the shipped text is the canonical English Rosenberg Self-Esteem Scale wording placed in an order inferred from the reverse-keying pattern in the raw responses; the study administered the Korean version (Jon, 1974), whose wording is not reproduced anywhere. What the data establish is each item’s polarity: S1, S2, S4, S6 and S7 intercorrelate positively (+0.31 to +0.73) and S3, S5, S9 and S10 form the opposing block, which rules out the scale’s original item numbering, and S8 is identifiable by its distinctive near-zero correlation with the positive block. What they do not establish is the order within each polarity block, so a swap between two same-polarity items – S1 and S2, say – is not excluded
baquerotomas_2026_gad7	Item wording is the official Spanish (US) GAD-7 form, chosen because the response labels in the deposit match that form exactly; the deposit itself records no item wording, and item order is assumed to follow the standard GAD-7 numbering.
baquerotomas_2026_phq9	Item wording is the official Spanish (US) PHQ-9 form, chosen because the response labels in the deposit match that form exactly; the deposit itself records no item wording, and item order is assumed to follow the standard PHQ-9 numbering.
bartoli_2022_badge_notifications	This table records a single-click behavioural task rather than a questionnaire. Each participant saw one home screen of 15 app icons with a badge notification on one app and made a single click, so resp = 100 marks the one app that participant clicked first and 0 the other fourteen. The response-option text is descriptive; the task offered no labelled options. The item text gives each app’s conventional name, with the Italian default-app labels left as administered
bate_2019_srq	Codes SRQ01-SRQ20 are matched to the paper’s Table 1 statements by row order, which the data support at the cluster level but cannot confirm item by item. All 20 items are stored reverse-scored where needed, so that higher always means better self-reported face recognition: for the ten negatively worded items (SRQ01, 03, 04, 05, 08, 10, 11, 12, 17, 19) resp 1 = strongly agree.
baumgaertner_2018_vaccine_trust	The item text is the study’s own variable labels (disease; low- vs high-risk scenario), not the wording respondents read, which was never published. Each scenario prompt contains only the fragment the paper quotes. A sixth option, ‘I don’t know’, was offered but is already missing in the source data, so it has no row.
beck_2021_iesr	resp values 1-4 are category indices, not scores. The study used the German IES-R (Maercker & Schuetzwohl, 1998), whose four anchors are ueberhaupt nicht / selten / manchmal / oft and whose traditional scoring weights that version prints are 0, 1, 3 and 5 – so summing resp gives a different scale from the published one. Item text is that German form verbatim, including its pre-1996 orthography
bfi_goldberg_1992_agreeableness	Despite the table name and its dictionary description (“Big Five Inventory”), these items are not the BFI-44. They are the Agreeableness scale of the 50-item IPIP Big-Five Factor Markers (Goldberg, 1992), as administered by the Open-Source Psychometrics Project.
bfi_goldberg_1992_conscientiousness	Despite the table name and its dictionary description (“Big Five Inventory”), these items are not the BFI-44. They are the Conscientiousness scale of the 50-item IPIP Big-Five Factor Markers (Goldberg, 1992), as administered by the Open-Source Psychometrics Project.
bfi_goldberg_1992_extraversion	Despite the table name and its dictionary description (“Big Five Inventory”), these items are not the BFI-44. They are the Extraversion scale of the 50-item IPIP Big-Five Factor Markers (Goldberg, 1992), as administered by the Open-Source Psychometrics Project.
bfi_goldberg_1992_neuroticism	Despite the table name and its dictionary description (“Big Five Inventory”), these items are not the BFI-44. They are the IPIP Big-Five Factor Markers (Goldberg, 1992) Emotional Stability scale, labelled here as Neuroticism, as administered by the Open-Source Psychometrics Project. Under that label, items N2 and N4 are the reverse-keyed ones, and they are stored unreversed.
bfi_goldberg_1992_openness_to_experience	Despite the table name and its dictionary description (“Big Five Inventory”), these items are not the BFI-44. They are the IPIP Big-Five Factor Markers (Goldberg, 1992) Intellect or Imagination scale, labelled here as Openness to Experience, as administered by the Open-Source Psychometrics Project.
bialowolski_2024_financial_literacy	Respondents were randomly assigned one of three wordings of each question (feminized, masculinized, gender-neutral Polish), so item_text lists all three; resp is scored 0/1, with ‘don’t know’ counted as incorrect, and the tie of item code FL_19 to the appendix’s 18th question is inferred from order (the deposit has no FL_18). The English item wording is the study’s own appendix translation of the Polish; only the two scoring labels (‘Incorrect answer’ / ‘Correct answer’) are IRW’s translation.
biswas_2024_digital_center_quality	The items were administered in Bangla, which is not published; the item text is the authors’ English. Item text is given for 22 of 26 items: the paper prints no wording for CONI4-CONI7, which are left blank. CONI3 is printed as a negatively worded item (‘would like to discontinue’) but correlates positively (r 0.29-0.65) with every other item, so either it was reverse-scored before deposit or the printed wording differs from what was administered. PAR1-PAR3 correspond to the paper’s CITP1-CITP3.
bitew_2020_lte	Item text is the study’s own abbreviated SPSS variable labels (“self illness”, “family death”), not the administered wording, which was in Amharic and was never published. Canonical Life Events Questionnaire wording was deliberately not substituted because this version departs from it: items 8-11 are reordered relative to the original sequence, and item 12 (intimate partner violence) is not an LTE-Q item at all
bitew_2020_osss3	Item wording is the canonical English OSSS-3 rather than the study’s administered Amharic version. One caveat on the mapping: item 1’s variable label in the source file reads “others worry for me”, which does not obviously correspond to the canonical item-1 content about how many people you can count on. It is identified as item 1 by its response structure instead – it is the only one of the three offering four categories rather than five, which is the instrument’s own signature, and the file’s total column equals the sum of the three items for every case
bitew_2020_phq9	Item wording is the canonical English PHQ-9; the study administered a locally validated Amharic translation whose exact wording is not published. The mapping to those canonical items is confirmed by the source file’s own numbered symptom labels and by the paper’s reported per-item means
bitew_2020_self_efficacy	This table holds 9 of the General Self-Efficacy Scale’s 10 items, and its item codes are offset by one: SEFFICAn is GSE item n+1, so SEFFICA1 is GSE item 2. The source file does contain GSE item 1, under a differently-named column, but it was not shipped to the IRW. The response anchors are the authors’ own shorthand for the GSE’s Not at all true…Exactly true scale
bled_2021_imagery_phenomenology	The item text is the administered French, with the paper’s own English alongside; the English instructions were produced by IRW. No separate response options were published, so the option text is taken from the alternatives named in each question, and the on-screen button labels are unknown. For pheno_detail the source data file labels 1 as detailed and 3 as blurry, but the paper’s reported percentages (66.7% of autistic vs 43.8% of control participants reporting detailed images) match resp = 3 exactly, so this table labels 3 as detailed and 1 as blurry.
boudreau_2022_police_reform	resp is not fully ordinal. Levels 1-4 run from strongly support to strongly oppose, but level 5 is “Don’t know”, which the questionnaire offered alongside the four substantive options and which about 11% of respondents chose on every item. Treating 5 as the oppose end of the scale will mis-order the data. Separately coded no-answer responses are dropped, as in the source.
BPAQ_Christopher_2024_PCL5	The paper describes the PCL-5 response scale as 0-4, but the deposited data (and this table) store it as 1-5, so the standard anchors are shipped shifted by one (1 = Not at all … 5 = Extremely); the paper’s reported PCL-5 total (M = 34.18, n = 109) is the 1-5 sum. Anchors and the instruction stem are the standard National Center for PTSD form, which the study’s own materials do not print.
brain_hemisphere	The instructions field describes the response scale rather than quoting an instruction shown to participants; the source dataset documents the scale but publishes no participant-facing instruction text. Only three of the five scale points were labelled (1 = Disagree, 3 = Neutral, 5 = Agree), so the option text is blank at 2 and 4. Note also that this 2015-16 dataset is the 20-item version of the scale; the version currently offered online has 24 items
brand_raffaelli_2024_familiarity_24	The numbered item codes are question positions in a counterbalanced design, not brands: each respondent was randomised to one of ten brand lists, so each code was answered about ten different brands, and the list assignment is not carried in the IRW table – items cannot be tied to a brand. The item text uses the brand-name survey’s wording; rows with item_family = logo were rated on a logo image and their prompt read ‘this logo’ rather than ‘this brand’.
brand_raffaelli_2024_liking_20	The numbered item codes are question positions in a counterbalanced design, not brands: each respondent was randomised to one of ten brand lists, so each code was answered about ten different brands, and the list assignment is not carried in the IRW table – items cannot be tied to a brand. The item text uses the brand wording; the two logo administrations read ‘this logo’. In addition, the logo Familiarity-Liking subsample’s responses (about 305 respondents) were mis-recoded during processing – values shifted one point, with ‘Dislike’ stored as 7 – so the 1 = Dislike / 7 = Like anchors shown do not describe those rows.
brand_raffaelli_2024_liking_24	The numbered item codes are question positions in a counterbalanced design, not brands: each respondent was randomised to one of ten brand lists, so each code was answered about ten different brands, and the list assignment is not carried in the IRW table – items cannot be tied to a brand. The item text uses the brand-name survey’s wording; rows with item_family = logo were rated on a logo image and their prompt read ‘like or dislike this logo’.
brand_raffaelli_2024_recognition_20	Old/new recognition memory for brand names and logos. An item code is a position in the Qualtrics loop, not a particular brand: the same code indexes a different serial position in the two subsamples (for item_family=name subtract one), and which brand or logo sits at that position depends on the respondent’s cov_condition. resp is 1 for ‘Old’ and 2 for ‘New’. Positions 1-50 of each subsample were the studied set for respondents in odd-numbered lists and the lures for even-numbered lists; positions 51-100 the reverse.
brederecke_2020_phq4	The item text is the study data file’s own terse English variable labels (“little interest or pleasure in your activities”), which are neither the German wording the instrument was administered in nor the canonical PHQ-4 phrasing. The response options are that file’s English labels too, and they likewise differ from the published PHQ-4 anchors – “On single days” where the instrument reads “Several days”. The instructions line is the canonical PHQ-4 stem, which the data file does not carry
brederecke_2020_sis	Items 4 and 9 are stored reverse-coded in the source data, and the option text here follows that stored direction: for those two items resp = 1 is “strongly agree” and 5 is “strongly disagree”, while the other nine run 1 = strongly disagree to 5 = strongly agree. This is stated by the study’s own file, which names those two columns “SIS Item 4 recoded” and “SIS Item 9 recoded” – they are the scale’s two negatively-worded items. item_text is the administered German from the paper’s Table 3, which is headed “Original item (German version)”, with the paper’s English in item_text_translated. The five-point anchors are an exception: neither the paper nor the study’s data file gives them in German – the file labels its values “1” to “5” with no anchor text – so option_text holds the English wording quoted above rather than what respondents read, and its _translated twin is empty
broadband_inventories	The instructions, quoted verbatim from the test’s own page, describe a five-point scale, but the items were administered on the seven-point Strongly disagree..Strongly agree scale that resp 1-7 encodes. The item wording is otherwise verbatim from the source codebook, including its typos (item 18 reads “when Im feeling badly”)
buczel_2022_inoculation_belief	Each of these six items covers six different question wordings. The study counterbalanced six scenarios across participants, so the sentence any one participant read depends on which scenario they saw; item_text therefore names the item’s condition and quotes one scenario’s wording as an example rather than presenting one participant’s version as the item. Note also that the response scale displayed to participants ran 0-10 while resp is stored as 1-11, so the option text carries the displayed number and is deliberately not equal to resp
bukurov_2022_sf36	Item and option wording is the public-domain English RAND 36-Item Health Survey. The study administered a Serbian-language SF-36, so this is canonical text rather than what participants read. The mapping to those canonical items is firm: the source file labels every column with its item, and the per-item response formats match the instrument’s mixed structure exactly (5 levels for items 1, 2, 20, 22 and 32-36; 3 for items 3-12; 2 for 13-19; 6 for 21 and 23-31)
burgess_2025_soas	The item codes are non-contiguous – SOAS5, 7, 9, 10, 11 and 12 – because they refer to positions in the 13-item version administered in this study, of which these six were retained in the final scale. The study’s data file carries no variable or value labels, so the item text and that numbering come from the paper’s own item table rather than from the data. In this study the items were read aloud to children who were shown a fixed object, so the published instrument’s opening line asking the respondent to choose and write down an object did not apply and is not included here
busch_2022_course_alleviate	These are the checkbox options of a single select-all-that-apply survey question, not independently administered items: resp = 1 means the respondent ticked that aspect and 0 means they did not, so a 0 records the absence of a tick rather than substantive disagreement. The question was shown only to the respondents who reported depression symptoms while taking online science courses
busch_2022_course_exacerbate	This battery is a select-all-that-apply checklist, so resp is a checkbox state rather than a chosen response option, and the option_text values ‘Selected’ and ‘Not selected’ describe that 0/1 coding rather than transcribing anything the survey printed – the source publishes no response labels for it, only the stem ‘Please select all aspects of ONLINE college science courses that might make your feelings of depression worse.’ Item text and instructions are verbatim from the study’s own S1 Appendix survey copy.
busch_2023_stigma	Despite the table name, these items are not stigma ratings. They are the participants’ own binary self-reports of concealable identities – LGBQ+, depression, anxiety, low SES, first-generation status, academic struggle, disability, community-college transfer – while the study’s actual 4-point perceived-stigma measure is not in the IRW table at all. Six of the eight items share the identical stem “I most closely identify as”, so the item content lives entirely in the response options. For two items the binary 0 collapses several source options (for example mid and upper income both map to 0), and the option text joins those source options verbatim; “Decline to state” responses are dropped rather than coded
butt_2022_actual_usage	This two-item scale does not use the questionnaire’s 1-7 agreement scale: each item has its own response anchors – a frequency scale for one and hours per week for the other – so instructions is blank because the block’s shared instruction does not apply
butt_2022_cognitive_absorption	The study’s data file contains four of the six Cognitive Absorption items – CA3 and CA5 were dropped by the authors – and it carries no variable labels for them, so the wording here comes from the appendix questionnaire, which lists the six items in order without codes. The alignment is therefore by position within that block
butt_2022_institutional_factors	Eleven of the appendix’s thirteen Institutional Factors items are in the data; IF3 and IF11 are absent from the study’s own file and are not shipped. The data file gives no variable labels for these columns, so the wording comes from the appendix questionnaire by position within the block. One source typo was corrected: the appendix prints “students ’questions” with a misplaced opening quote, shipped here as “students’ questions”
buzgova_2023_gai	Item text is the canonical English Geriatric Anxiety Inventory; the study was Czech and its translation is not published, and the data file carries no labels of any kind. The mapping of code to item is supported but not settled: the instrument’s published three-factor structure reproduces for 15 of the 20 items, with the three purely somatic items all behaving as expected, while five cognitive-versus-arousal items are ambiguous
buzgova_2023_gds	Item text is the canonical English GDS-15; the study was Czech and its translation is not published in the paper or the data file, which carries no labels of any kind. The responses are stored already scored rather than raw, so resp = 1 always marks the depression-consistent answer and the option text follows that stored direction rather than the yes/no order printed on the questionnaire
buzgova_2023_lsita	This table is the 12-item short form of the LSITA, and the item text is the canonical English wording – the study was Czech and its translation is published nowhere accessible, the data file carries no labels, and the companion validation paper is paywalled. Items 2, 4, 5 and 6 are stored already reversed, so their option text is flipped relative to the printed instrument; the raw column sum reproduces the paper’s published total exactly, which is what establishes that
buzgova_2023_qol	The OPQOL-brief was administered in Czech, but the Czech wording is not published in the study’s paper or data file, so the canonical English OPQOL-brief items are shown. Responses are stored on the form’s raw coding (1 = Strongly agree .. 5 = Strongly disagree), so for every item a lower value means better quality of life, although the paper describes higher scores as better. Source credit, as requested by the copyright holder: OPQOL-brief, copyright A. Bowling; Bowling A, Hankins M, Windle G, Bilotta C, Grant R (2013), Archives of Gerontology and Geriatrics 56(1):181-187, doi:10.1016/j.archger.2012.08.012.
buzgova_2023_rses	The item numbering here is an assumption that could not be independently confirmed. The study’s data file carries no variable or value labels of any kind and the paper reproduces no items, so the text is the canonical English Rosenberg Self-Esteem Scale mapped on the basis that RSESn is Rosenberg item n. That assumption is known to fail for this instrument: in bakker_2020_rses the columns were numbered in canonical order while the administered form was permuted. Here the responses are stored already reverse-scored – every inter-item correlation is positive – so the keying-polarity check that caught the other case cannot be run. One position does have positive support: RSES8 has both the lowest mean and the lowest item-rest correlation, the known signature of Rosenberg item 8 (“I wish I could have more respect for myself”). Positions 1-7, 9 and 10 rest on the numbering alone. The scale is also stored pre-reversed, so the same anchors apply to every item. The study administered a Czech translation that is not reproduced in the paper or the data file, so this English is not what respondents read
buzgova_2023_soc	Item text is the canonical English SOC-13; the study was Czech and its translation is not published in the paper or the data file, which carries no labels of any kind. The responses are stored already reverse-scored, so the option text for the five reverse-worded items (1, 2, 3, 7 and 10) is flipped relative to the printed instrument. Note also a discrepancy with the paper: it reports Cronbach’s alpha of 0.92 for this scale, while the released data give 0.765 with corrected item-total correlations of 0.29-0.53
buzzi_2025_autonomy	The English translations (instructions, item and option text) were written by IRW from the Italian questionnaire respondents read; they are not the study’s own.
c19prc_uk_mcbride_2021_drinkcontext	resp means different things by wave: for the _W2 items 1-4 run Never..Always, while for the _W3 items 0=No and 1=Yes, so resp=1 is Never at wave 2 but Yes at wave 3.
c19prc_uk_mcbride_2021_homespace	For the two child items (Home3, Home4) roughly 40% of responses sit at the 4 = Neither agree nor disagree midpoint, and most of those come from respondents who reported no children in the household, so the midpoint there largely reflects not-applicable answers rather than neutral agreement.
c19prc_uk_mcbride_2021_hygiene	Items Hygiene1-14 pool waves 1 and 2, and the shipped stem is the wave 1 wording; at wave 2 the example list in the stem additionally read ‘avoiding touching eyes and mouth, using tissues’.
c19prc_uk_mcbride_2021_infosource	The lowest response option was labelled ‘None’ in waves 1-2 and ‘Not at all’ in waves 3-4; option_text for resp=1 gives both as ‘None/Not at all’.
c19prc_uk_mcbride_2021_ius12	Item codes IOU1-IOU12 follow this study’s own item order (prospective items first), not the standard IUS-12 numbering of Carleton et al. (2007); and the midpoint label ‘Somewhat characteristic of me’ (resp 3) appeared on the wave 1 questionnaire only, the wave 2 questionnaire showing that point as a bare ‘3’.
c19prc_uk_mcbride_2021_lockdown_contact_behaviours	Item codes follow the wave 5 questionnaire’s numbering. Risk_Behaviours_5 and Risk_Behaviours_6 were asked at wave 5 only; wave 6 dropped them and renumbered the rest, so the wave 6 questionnaire’s items 5-7 are stored as Risk_Behaviours_7-9.
c19prc_uk_mcbride_2021_lockdownbehaviour	Items Risk_Behaviour1-7 pool waves 2 and 3 and the shipped wording is wave 2’s; at wave 3 the stem read ‘In the past week, on how many days have you:’ (without ‘to reduce your risk of being infected by or passing on the coronavirus to others’) and several behaviours were lightly reworded, while items 8-16 were asked at wave 2 only.
c19prc_uk_mcbride_2021_mabs	Response codes run 1 = Strongly agree to 5 = Strongly disagree, so higher values mean less agreement; MABS_god_aware was asked at wave 4 only, and wave 4 moved ‘The soul is immortal’ from first to last position (item codes follow content, not position).
c19prc_uk_mcbride_2021_phq15	The study administered 14 of the 15 PHQ-15 items (the menstrual item was omitted) and numbered them consecutively, so som_4..som_14 correspond to canonical PHQ-15 items 5..15, not items 4..14; the timeframe is the past 7 days (APA DSM-5 Level 2 adaptation) rather than the original PHQ-15’s past 4 weeks.
c19prc_uk_mcbride_2021_phq9	In waves 1 and 2 (about a quarter of Dep_9 responses) item 9 read ‘Thoughts that you would be better dead or of hurting yourself in some way’; waves 3-6 used the shipped wording ‘…better off dead…’.
c19prc_uk_mcbride_2021_polarization_brexit	Despite the study’s label ‘affective polarization’, these four items ask how much events connected to Brexit affected the respondent’s mental health or made them anxious, angry or distressed; they do not measure attitudes toward the other Brexit side.
c19prc_uk_mcbride_2021_protect	Responses of ‘Not applicable to me (I had no plans to do this)’ were removed from the response data, so each item’s n reflects only respondents who had plans to do the activity (from 770 for travelling to infected areas to 1,961 for touching eyes or mouth).
c19prc_uk_mcbride_2021_protective_behaviours	Items pool waves 5 and 6 and the shipped wording is wave 5’s; at wave 6 the stem read ‘To reduce your risk of being infected by COVID-19 or to reduce the spread of COVID-19 in the community, in the past week, how often have you:’ and Hygiene_12 read ‘Stayed at least 1 metre (3ft) away’ instead of ‘2 metres (6ft)’, so Hygiene_12 responses (2058 of 4578) at wave 6 refer to a 1-metre threshold.
c19prc_uk_mcbride_2021_riskbehaviour_c19	Items C19_Risk1-5 pool Waves 1-3: Wave 1-2 respondents were asked ‘To reduce your risk of being infected by the coronavirus COVID-19 have you recently…’, while the 2,019 Wave 3 responses per item were given under ‘To reduce your risk of being infected by COVID-19, in the past week have you…’.
c19prc_uk_mcbride_2021_riskreduction_beliefs	RiskReduction1’s text reproduces the Wave 1 misspelling ‘Antibotics’ that Wave 1 respondents saw; Wave 2 respondents (1406 of 3431 responses) read ‘Antibiotics are effective in preventing and treating COVID-19’.
c19prc_uk_mcbride_2021_socialdistance	SocialDistance1-14 pool waves 1, 2 and 5 (SocialDistance4 waves 1 and 2 only), and SocialDistance15_W2-17_W2 and SocialDistance18 pool waves 2 and 5; the shipped stem is the earliest wave’s. The stem’s examples changed by wave: wave 2 read ‘social distancing rules (e.g. staying in your home most of the time, exercising outside once a day, not meeting up with friends and relatives, maintaining a 2-metre distance from people, working from home)’, wave 5 ‘social distancing practices (e.g. not entering other people’s homes, maintaining a 2 metre distance from people, working from home, etc)’. Wave 5 numbers its items differently and presented SocialDistance18 eighth.
c19prc_uk_mcbride_2021_trustinstitutions	Two institutions were worded differently across waves – waves 1-3 read ‘Parliament’ and ‘The government’ where waves 4-5 (shipped) read ‘The UK Parliament’ and ‘The UK Government’ – and the wave 3 prompt read ‘Please indicate how much trust you have in the following institutions?’; resp is harmonised so 5 = Completely trust in every wave, although waves 1, 2, 4 and 5 were stored in the opposite direction in the source files.
c19prc_uk_mcbride_2021_trustsource	The third response option was labelled ‘Somewhat’ in waves 1-2 (shipped) and ‘Some’ in waves 3-4.
cacciatore_2021_crisis_care_satisfaction	The resp values are not the survey’s own choice numbering. The questionnaire numbered its options 1 = Extremely satisfied down to 6 = N/A, while the IRW values are the processing script’s recoding into ascending satisfaction, and the “N/A” responses are dropped rather than kept as a level. Item text is the matrix row label and the shared stem is recorded once in instructions; respondents rated only the care providers they actually encountered, which is why the number of responses varies widely across items
campos_2023_oes	This table pools Finnish-language (n=3,614) and Brazilian Portuguese-language (n=3,979) administrations of the Orofacial Esthetic Scale, and the item text is the English source instrument (Larsson et al. 2010, as reprinted by Campos et al. 2020) rather than either administered version; no file in the study’s deposit or supplements gives the items in any language. The item-to-code mapping is reconstructed rather than stated by the study. The shipped English instructions include a clause about tooth replacements (crowns, bridges and implants) that the Portuguese instruction omits, so the administered framing was not identical across the two halves.
campos_2023_pidaq	This table pools Finnish-language (n=3,614) and Brazilian Portuguese-language (n=3,979) administrations of the 24-item PIDAQ, and the item text is the English source instrument (Klages et al. 2006, as reprinted by Campos et al. 2020) rather than either administered version; no file in the study’s deposit or supplements gives the items in any language. The English is not a word-for-word match to the Portuguese version for every item (it9, it16 and it22 differ in content), and the Finnish administration reworded item 22 to be gender-neutral while the original wording is shipped. Item codes follow the original interleaved PIDAQ numbering, not the subscale-sequential renumbering some validation papers use.
carney_2023_substance_use	The eight items are indicators the study derived, not questions respondents answered: 1 means any use of that substance in the past three months on the WHO ASSIST, except alcohol, where 1 means an AUDIT score above 8. The paper reproduces no ASSIST or AUDIT wording, so item_text describes each indicator in the paper’s words and option_text gives the thresholds, not administered anchors.
carver_2017_puggs_pilot1_attitudes	The item and option wording is the authors’ British-English original of the first-pilot PUGGS questionnaire (S3 Table), but the first pilot was administered to Brazilian students in a Brazilian Portuguese translation that none of the paper’s supplements publishes, so this is not the wording respondents read. The questionnaire’s fifth option, ‘don’t know’, is dropped from the response data, so the table has only the four agreement levels.
carver_2017_puggs_pilot1_det_core	The study’s deposit publishes these 13 statements twice, under two different numberings: the pilot-1 questionnaire and the pilot-1 codebook order them differently, and only 3 of the 13 positions agree. The released data follow the codebook, so the item text here is the codebook’s wording and numbering; it differs in minor wording from the questionnaire respondents were handed, and for some items may be a later revision of it. Both pilots were administered to Brazilian students in Brazilian Portuguese and no Portuguese wording appears anywhere in the deposit, so the text shown is the authors’ English original and the _translated columns are empty – the standard’s signal that the base fields are not what respondents read.
carver_2017_puggs_pilot1_genom_know	Despite the table name, resp here is not a correctness score: it is the raw 4-point agreement scale the first pilot administered (1 = strongly disagree to 4 = strongly agree), so option_text is meaningful. correct_response accordingly holds the agreement levels the study’s codebook counts as understanding – “3;4” where the statement is true, “1;2” where it is false – rather than a single keyed option. “Don’t know” was an offered response that IRW processing drops as non-response rather than scoring it wrong, so per-item n varies with item difficulty and is lowest on the five statements that mention epigenetics. As with carver_2017_puggs_pilot1_det_core, the deposit numbers these 18 statements two different ways and the shipped text follows the codebook, which is the numbering the data demonstrably use. The pilot was administered in Brazilian Portuguese and the deposit publishes no Portuguese wording, so the English original ships with the _translated columns empty.
carver_2017_puggs_pilot1_traits	This is the first pilot’s table-of-traits matrix: 20 traits, each rated 1-5, with the shared stem and rubric recorded once in instructions rather than repeated into every item. Respondents could also mark a sixth “don’t know” column, which IRW processing drops, so it has no resp level here. The questionnaire’s worked example row is not an item and is excluded. correct_response is blank: the codebook publishes an expected answer per trait, but it is a scientific reference value for a belief item rather than a response key. The pilot was administered to Brazilian students in Brazilian Portuguese and the deposit publishes only English, so the item and option text is the authors’ English original and the _translated columns are empty – the standard’s signal that the base fields are not the wording respondents read.
carver_2017_puggs_pilot2_attitudes	Six of these 20 items are negatively worded and the study’s codebook reverses them before scoring; they are stored raw here, so anyone summing or averaging the scale must reverse them first. resp is the codebook’s primary coding as administered (1 = strongly disagree to 4 = strongly agree), not its secondary “very negative to very positive” recode. The second pilot was administered to Brazilian students in Brazilian Portuguese and the deposit publishes no Portuguese wording, so the item and option text is the authors’ English original and the _translated columns are empty.
carver_2017_puggs_pilot2_det_core	resp is a correctness score, not the answer the respondent gave: participants chose True, False or Don’t know, and the study’s own key scores the keyed answer 1 and the other 0. IRW processing drops “Don’t know” and missing as non-response rather than scoring them incorrect – narrower than the study’s own codebook, which scores don’t-know as 0 – so a 0 here always means a wrong True/False answer, and “Don’t know” has no resp level of its own. option_text names which literal choice each score corresponds to on each item. The second pilot was administered to Brazilian students in Brazilian Portuguese and the deposit publishes no Portuguese wording, so the text is the authors’ English original and the _translated columns are empty.
carver_2017_puggs_pilot2_genom_know	resp is not the response the student gave but a correctness score: each of these 16 statements was asked as True / False / Don’t know, and IRW processing records 1 where the answer matches the study’s own published key and 0 where it does not. option_text is therefore blank throughout by design, and correct_response holds the keyed answer (“True” or “False”) rather than a resp value. Don’t-know and missing responses are dropped as non-response rather than scored incorrect – narrower than the study’s own codebook – so per-item n runs from 35 to 70 of the 78 respondents and is lowest on the five statements about epigenetics, which the paper itself reports drew the most don’t-knows. The items were administered in Brazilian Portuguese and the deposit publishes only English, which is what ships here, with the _translated columns empty.
carver_2017_puggs_pilot2_traits	This is the second pilot’s table-of-traits matrix: 17 traits, each rated 1-5, with the shared stem in instructions rather than repeated into every item. It is a different trait list from the first pilot’s, which has 20 and includes a sixth “don’t know” column this form does not. correct_response is blank, since the pilot-2 codebook publishes no expected answer for the trait matrix. The pilot was administered to Brazilian students in Brazilian Portuguese and the deposit publishes only English, so the item and option text is the authors’ English original and the _translated columns are empty – the standard’s signal that the base fields are not the wording respondents read.
cavalini_mokken	Seventeen ways of coping with a bad smell from a nearby factory, each rated 0 = never to 3 = always. The study was conducted in Dutch; the English shipped here is the mokken R package’s rendering, which is the form in which this data set is distributed. The seventeen items form four scales (Sijtsma & Molenaar 2002, pp. 82-86).
ccapsvtskhpacr_mercedes_2023_beck	The item text here is the administered Spanish, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. The official English BDI-II wording is Pearson-licensed, so the translated column is IRW’s own rendering of the Spanish statements the study published, not Beck’s English. Compound options where the source joins two statements with ; are kept as one string in both columns so the two stay row-aligned.
celik_2026_academic_motivation	The item text is the Turkish wording from the study’s own online form, which matches the published Turkish Academic Motivation Scale (Ünal-Karagüven, 2012) item for item. The deposit stores bare integers 1-7, so the labels for points 1, 4 and 7 (‘Hiç uyuşmuyor’, ‘Orta derecede Uyuşuyor’, ‘Tam olarak Uyuşuyor’) come from that published form, and it is not confirmed that the study’s form showed them; points 2, 3, 5 and 6 are unlabelled. The English in the _translated columns is Vallerand’s own AMS-C 28 English, a parallel version of the same numbered items rather than a translation of the Turkish, except that the ‘Because …’ in the instructions is IRW’s rendering of ‘Çünkü …’.
celik_2026_tipi	The item text is the Turkish wording from the study’s own online form, which matches Atak’s (2013) Turkish Ten-Item Personality Inventory. The deposit stores bare integers 1-7, so the seven scale labels (1 = ‘Tamamen katılmıyorum’ … 7 = ‘Tamamen katılıyorum’) come from Atak’s published form, and it is not confirmed that the study’s form showed them. The English in the _translated columns is Gosling et al.’s (2003) TIPI English, a parallel version of the same numbered items rather than a translation of the Turkish. Responses are stored raw: the reverse-keyed TIPI items (2, 4, 6, 8 and 10) are not reversed.
cfq_ruiz_2025_dass	Despite the table name, this is the study’s DASS-21, not the Cognitive Fusion Questionnaire. The administered Spanish is the deposit’s own form, a hybrid of two published Spanish versions; the English columns are the official English DASS-21 matched by item number, a parallel form rather than a literal translation of this Spanish (item 18 is one example). One form was deposited for all five countries sampled. Subscale membership is confirmed for 19 of 21 items; order within each subscale follows the questionnaire’s numbering.
chatton2024_honos13	The HoNOS-13 was administered in French; no French wording was deposited, so the English item names printed in the source paper’s Table 1 are shipped in their place, and the 0-4 response anchors come from the Royal College of Psychiatrists’ HoNOS glossary rather than from the study, which labels only the two endpoints.
chen_2021_acculturation	The BAS-MBIW was administered orally in Mandarin Chinese; the item wording shown is the authors’ own published English, as no Chinese-language version appears in the article or its single supplementary file. Note also that responses are stored 1-5 while the paper scores the scale 0-4, and that the paper’s Table 1 names the anchors ‘rarely…usually’ where the data file’s own value labels read ‘never…always’ (the latter are shipped).
chen_2021_enculturation	The BAS-MBIW was administered as a Mandarin Chinese interview (with an interpreter where needed), but no Chinese wording is published in the paper or its S1 Data file, so the item text shown is the English wording carried in the data file’s own SPSS variable labels. Note also that the published scale is scored 0-4 (never..always) while the IRW table stores 1-5, so the anchors here are attached to the shifted coding.
chen_2022_gad	The deposit does not include the Chinese GAD-7 respondents actually read, so the item and response-option wording is the standard English GAD-7 (Spitzer et al. 2006) as distributed on phqscreeners.com, standing in for it. In the source data item GAD03 has no responses at all in its top category (‘Nearly every day’, 0 of 77,211), which probably reflects a collection or coding problem in the original data.
chen_2022_sasc	Despite the table name, SASC here is the Smartphone Addiction Scale for College Students (SAS-C; Su et al., 2014), measuring problematic smartphone use – not the Social Anxiety Scale for Children. The survey was administered in Chinese and the wording shown is the depositors’ own English item labels from the SPSS file, as no Chinese original is published with the dataset
chen2022_cls	Item wording is the canonical English CLS and refers to the school context, while the study administered a Chinese version and interprets the scale as loneliness at home; the six reverse-worded items (A1, A4, A8, A10, A16, A22) are stored already reverse-scored, so their response labels run in descending order (resp 1 = ‘Always true’).
chen2022_ses	The Rosenberg Self-Esteem Scale wording here is the canonical English original (Rosenberg 1965); the study administered a Chinese adaptation (Ji and Yu, 1999) whose wording it does not publish. The mapping of items to the file’s bare codes A25-A34 is inferred from questionnaire position – no source ties an individual code to an individual item, and nothing distinguishes A27 from A28. Response options are blank because the deposited values are already keyed toward high self-esteem rather than to the paper’s stated ‘1 = very conformity … 4 = very inconformity’ anchors, and A34 runs opposite to the other nine (item-rest r = -0.16).
chen2022b_psychabuse	Item text is the authors’ English rendering copied from their data file’s column headers (typos kept; the Chinese questionnaire wording is unpublished), and the response anchors are shipped as 1 = very inconformity and 5 = very conformity, reversed from the paper’s printed ‘0 = very conformity to 4 = very inconformity’ because the stored codes run higher = more abuse, as the paper’s own means and prevalence figures confirm.
chen2022b_selfesteem	Responses in this table are stored already direction-aligned (higher = higher self-esteem), so the 1-4 codes do not share one anchor orientation across the ten items; the paper’s stated anchor direction is contradicted by its own data, so no response-option wording is shipped. Item text is the authors’ English rendering of the items, taken verbatim from their data file’s column headers (source typos retained); the instrument was administered in Chinese and no Chinese wording is published in the article or the deposit.
chen2022b_socsupport	Item text is the canonical English MSPSS wording; the study administered a Chinese version and no Chinese wording is published in the paper or its data deposit, and the anchors for scale points 2-6 are the canonical MSPSS labels rather than ones the study printed.
chile_2023_children-adolescents-survey_aa	The English in the _translated columns of this table was produced by IRW, not by the study: EANNA 2023 published its questionnaire only in Spanish, and the Spanish in the base fields is what respondents actually heard. Note also that ten of the 38 items (a8_78, a8_87, a8_88, a8_99, a9_78, a9_87, a9_88, a9_99, a11_88, a11_99) are the ‘nobody’, ‘not applicable’, ‘don’t know’ and ‘no answer’ checkboxes of three multiple-select questions rather than substantive items
chile_2023_children-adolescents-survey_cp_a	The English in the _translated columns of this table was produced by IRW, not by the study: EANNA 2023 published its questionnaire only in Spanish, and the Spanish in the base fields is what respondents heard. Note also that these are household-level questions but the table has one row per selected child’s caregiver, so the 17,659 ids come from 13,867 households and the answers of 3,792 ids repeat another id’s household exactly
chile_2023_children-adolescents-survey_cp_c	The English in the _translated columns was produced by IRW, not by the study: EANNA 2023 published this questionnaire only in Spanish. Two defects in the response data itself: in cp9_1-cp9_4 the value 89.5 is the survey’s ‘don’t know’ code (entered as 88:88, converted to hours), not a number of hours, and should be treated as missing; and in cp2_1-cp2_9 (reasons for starting work) 3,493 of 8,793 caregivers have all nine reasons marked, a pattern already present in the source deposit that looks like a coding fault, so those nine items should not be analysed as recorded
chile_2023_children-adolescents-survey_g	The English in the _translated columns of this table was produced by IRW, not by the study: EANNA 2023 published its questionnaire only in Spanish, and the Spanish in the base fields is what respondents read. The eight items are the bullying sub-scale of the Escala de Clima Social Escolar (ECLIS; Aron, Milicic & Armijo, 2012), stored raw as 1=Nunca to 4=Siempre with no reverse-scoring applied (g1_3 is positively worded)
chile_2023_children-adolescents-survey_n	The English in the translated columns of this table was produced by IRW, not by the study: EANNA 2023 published its questionnaire only in Spanish, and the Spanish in the base fields is what the children (aged 5-8) heard from the interviewer. Response codes run in different directions across items (n5 1=Mal to 3=Bien, but n6 1=Mucho to 3=Nada and n9 1=Muy contento to 3=Nada contento), n10 is a nominal who-does-the-housework item, and n7/n8_ are 0/1 dummies of a multiple-select question that include non-substantive alternatives (A nadie, No aplica, No sabe, No responde)
chile_2023_social-welfare-survey_a	Fifteen subjective-wellbeing questions from Módulo A of Chile’s 2023 Encuesta de Bienestar Social (life satisfaction, yesterday’s happiness, calm, anger and sadness, and satisfaction with ten life domains), in the Spanish of the official questionnaire with IRW’s own English translation alongside – treat the Spanish as authoritative. a4 and a5 were asked only of employed respondents. Verified by reproducing all 75 of the codebook’s published response frequencies exactly from the live data.
chile_2023_social-welfare-survey_ee	The item text here is the administered Spanish, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. Item texts are the survey’s own abbreviated field labels, truncated in the source and left verbatim.
chile_2023_social-welfare-survey_f	The item text here is the administered Spanish, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. Item texts are the survey’s own abbreviated field labels. The source abbreviates to fit a column width (Ult. 12 meses, org., RRSS) and contains stray zero-width characters; both are kept as published.
chile_2023_social-welfare-survey_g	The item text here is the administered Spanish, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. Item texts are the survey’s own abbreviated field labels, truncated in the source and left verbatim.
chile_2023_social-welfare-survey_h	The item text here is the administered Spanish, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. Item texts are the survey’s own abbreviated field labels. NNA (ninos, ninas y adolescentes) is expanded in the translation.
chile_2023_social-welfare-survey_oo	The item text here is the administered Spanish, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. Item texts are the survey’s own abbreviated field labels. Parenthetical population markers (Ocupados, Cuenta Propia, Desoc, FFT) are expanded in the translation.
chile_2023_social-welfare-survey_rr	The item text here is the administered Spanish, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. Item texts are the survey’s own abbreviated codebook labels (e.g. fam., serv., RRSS), which the translation expands.
chile_2023_social-welfare-survey_ss	The item text here is the administered Spanish, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. Item texts are the survey’s own abbreviated codebook labels, which the translation expands.
chile_2023_social-welfare-survey_u	The item text here is the administered Spanish, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. Item texts are the survey’s own abbreviated codebook labels, which the translation expands.
chile_2023_social-welfare-survey_vv	Five housing-quality questions from Módulo V (Vivienda) of Chile’s 2023 Encuesta de Bienestar Social, in the Spanish of the official questionnaire with IRW’s own English translation alongside – treat the Spanish as authoritative. vv1_a-vv1_c answer the shared question ‘Su vivienda, ¿presenta alguna de las siguientes situaciones?’ (does your home present any of the following situations?); a fourth sub-item of that grid, vv1_d (noise), is not in this table. The response table also holds 20 rows with a missing response, which are the survey’s don’t-know/no-answer codes.
chile_2023_social-welfare-survey_yy	Six items from Módulo Y: Ingresos (income) of Chile’s 2023 Encuesta de Bienestar Social, in Spanish from the official questionnaire, with an English translation that is IRW’s own – treat the Spanish as authoritative. yy3 is not a rating item: it asks how much monthly income would be enough to cover the household’s needs, answered as an amount in Chilean pesos (10,797 answers, 60,000 to 14,000,000, heaped at round figures), so it carries no option text and should not be analysed alongside the 2-5 point items without treatment. Verified by reproducing all 18 of the codebook’s published response frequencies for the categorical items exactly from the live data.
chinvararak_2021_phq15	Item text and instructions are the PHQ-15’s canonical English wording, not the Thai the 180 respondents actually read: the PLOS deposit and supplements contain only 2-4 word Thai keyword tags as SPSS variable labels, never the Thai stems, so language is set to Thai beside an empty item_text_translated to mark the table for a later backfill. The response options are genuine administered Thai (ไม่รบกวน / รบกวนเล็กน้อย / รบกวนมาก = 0/1/2), taken from the deposit’s own value labels.
choy_2022_extraneous_events	The four item wordings come from the paper’s Table 1, which prints them as an unnumbered list, so their assignment to the codes Extraneous1-Extraneous4 rests on presentation order; a CFA on the live data reproduces the paper’s published Extraneous loadings (.67/.77/.82/.85 vs .67/.76/.83/.85) and confirms the codes match the paper’s item numbering, but not that the list is in that order.
christensen_2018_wsssf_2171	resp is the SCORED value, not the literal answer: 1 always means the schizotypy-keyed response, and option_text records which literal True/False answer that was for each item, because the keying direction differs item by item (e.g. resp=1 is ‘True’ for pb01 but ‘False’ for py01). Item codes mi/pb/py/sa do not encode item content.
christensen_2018_wsssf_5831	option_text is item-dependent for this table, because it stores keyed scores rather than raw answers: resp=1 reads ‘True’ on the 42 items the study does not mark reversed and ‘False’ on the 18 it does, all items having been scored so that a higher score means more schizotypy. The endorsement rates confirm it (the reverse-keyed items are ordinary pleasant experiences and run 0.036-0.276, not the ~0.9 a raw true/false coding would give). The authors’ ’ (reversed)’ annotation was stripped from item_text as scoring notation rather than administered wording, and no instruction text is published anywhere in the deposit or article, so instructions is empty.
chuemchit_2024_nonpartner_violence	Each of the five items is an author-derived binary indicator summarising several separate questionnaire questions, not one question: the item_text shipped here is the authors’ own definition of that violence type (from the paper’s Methods), and the study publishes no key saying which of the 14 non-partner questions (S1 File Q126-Q191) feed each indicator. The interviews were conducted in Khmer, Lao and Myanmar; only the English questionnaire version is deposited, so the shipped wording is not the string respondents heard.
chuemchit_2024_partner_violence	The five items are derived binary indicators, each collapsing 2-4 questionnaire questions of one violence type crossed with a ‘in the last 12 months’ follow-up, so item_text quotes the constituent questions rather than a single stem. The survey was administered orally in Lao, Burmese and Khmer and only the English master questionnaire was published, so the shipped wording is an English substitute for what respondents heard.
cinar_tanriverdi_2023_gad7	item_text ships the Turkish wording of the GAD-7 adaptation the study used (Konkan et al. 2013). The English in the _translated columns is Pfizer’s official GAD-7 wording for the same numbered items, not a word-for-word translation of that Turkish – the two diverge, for example Turkish item 1 is phrased as a direct question and item 6 adds a restlessness clause. No administered instruction line is shipped because none of the study’s sources print one.
clemente_2024_attitude_father	The deposit labels FATHER9 ‘Does not see child’, but item 9 of the published questionnaire reads ‘Creo que no quiere a sus hijos’ (I believe that he/she does not love his/her children); the item text follows the questionnaire.
clemente_2024_attitude_mother	The deposit labels MOTHER9 ‘Does not see child’, but item 9 of the published questionnaire reads ‘Creo que no quiere a sus hijos’ (I believe that he/she does not love his/her children); the item text follows the questionnaire.
clemente_2024_pmd	The English in the _translated fields is IRW’s translation of the administered Spanish wording, not the published English PMD items.
clemente_2024_sd4	The English in the _translated fields is IRW’s translation of the administered Spanish wording, not the published English SD4 items.
climatechange_geiger_2025	Respondents in Brazil, China, Germany, India, Indonesia, Italy, Japan, Mexico, Poland, Thailand and French-speaking Canada answered translated versions; the item text shown is the English (Canada) administration, with [citizens] and [country] standing for the nationality and country name piped in per country. Every translation is in the study’s Qualtrics files at osf.io/6hny4. Note also that own.ccb is stored 0-4 (the authors’ preparation script subtracts 1), although the study codebook describes it as 1-5.
clipa_2025_mslq	Respondents answered on a 1-5 scale (the original MSLQ uses 1-7) whose anchor labels the study did not publish, so response options carry no text. The eight negatively worded items (VAR00033, 37, 40, 52, 57, 60, 77, 80) are stored raw, not reverse-scored; this reproduces the deposit’s own 15 stored subscale means.
COACH_Chen_2022_ADL	The COACH trial was conducted in Chinese, but no Chinese wording is deposited; the item and response text shown is the study codebook’s own English rendering. Item ra_adl_q9 is labelled ‘Washing’ in that codebook, but respondents report more difficulty with it than with ‘Bathing’ (ra_adl_q10), so the underlying item is likely doing laundry rather than personal washing.
COACH_Chen_2022_CSQ	This table is dictionary-described as the 8-item Client Satisfaction Questionnaire (CSQ-8), but its items are not that instrument: they are items 1-8 of Baker’s Consultation Satisfaction Questionnaire short form (CSQ-9), rated 1 ‘Strongly agree’ to 5 ‘Strongly disagree’, taken verbatim from the study’s own codebook. The two reverse-worded items (q3 and q8) are stored raw, and the wording shipped is the deposit’s English codebook text – the questionnaire was administered in Chinese, and no Chinese wording appears anywhere in the deposit.
COACH_Chen_2022_IADL	Item and option wording is the English text of the study’s own codebook; the trial was administered in Chinese and the administered wording is not published in the deposit or the paper. The eight IADL items deliberately have different numbers of response options (three to five), which is how the Lawton IADL is written, so the shipped option text is keyed correctly item by item. Separately, 65 of the table’s 54,653 responses (0.12%) hold a value above their own item’s maximum and therefore carry no option text; these occur only at the 6- and 12-month waves, and at baseline every item’s largest observed value equals its expected ceiling.
COACH_Chen_2022_MOS_SSS_C	The 19 MOS Social Support Survey items are shipped against the questionnaire’s own numbering (Sherbourne & Stewart 1991) rather than the study codebook’s. That codebook lists the items grouped by subscale and numbers them Q2-Q20 in that grouped order, but the responses match the canonical numbering: the four tangible-support items are q2, q5, q12 and q15, which load 0.70-0.81 on their own factor and are the four highest-mean items in all three waves (3.96/3.93/3.91/3.76 against 2.62-3.04), where the codebook would place them at q10-q13. The order of items within a subscale is not independently confirmed. The wording shown is the English original; the survey was administered in Chinese and no Chinese text appears anywhere in the deposit.
COACH_Chen_2022_treatmentStigma	The wording and Yes/No labels are the study codebook’s English; the trial was administered in Chinese, and no Chinese wording is published. resp=1 means ‘Yes’ for item 1 but ‘No’ for items 2 and 3, so 1 is the stigmatising answer throughout. The baseline wave (wave 1) looks unreliable: every one of 2,365 respondents answered 1 to item 3, and baseline answers are essentially uncorrelated with each other (r = 0.09 for items 1-2, against 0.57 at 12 months) and with the same person’s 12-month answers.
codella_2020_school_efficacy	The item text is the administered Italian, taken from the deposited spreadsheet’s own final row. The 1-5 response anchors (“Totally unable” … “Totally capable”) and the shared stem the items complete (“How well can you”) are shipped in ENGLISH, because that is the only language the source publishes them in: the article states the anchors in English in its Measures section and the supplementary tables print the stem in English, while neither the deposit nor the article gives the Italian for either. So a single table mixes the administered wording of its items with an English rendering of its scale, and the anchors are not the words the children read.
cogcontrol_gyurkovics_2019_flanker	Item text for this table was reconstructed by IRW, not transcribed: the OSF deposit identifies each stimulus only by a numeric image id and ships no stimulus images, so the arrow display and correct keypad key shown for each item were decoded from the trial-level data (two independent columns agreeing across all 118 participants); the arrows are a rendering of images the deposit does not include, and no verbatim participant instructions are published.
cogcontrol_gyurkovics_2019_sart	The item text for this table is the stimulus digit itself, which is all a SART trial presents; no verbatim participant instructions are published by the study, so the instructions field is empty, and option_text holds the deposit’s accuracy codes (0 incorrect / 1 correct) rather than response options participants read.
cogcontrol_gyurkovics_2019_simon	Item text for this table was reconstructed by IRW, not transcribed: the OSF deposit identifies each Simon stimulus only by a numeric image id and ships no stimulus images, so the arrow direction, its screen location and the correct keypad key shown for each item were decoded from the trial-level data (correct-trial key code, 100% pure across 34,338 trials, plus the congruency column and the deposit’s own feature-repetition code); the arrows are a rendering of images the deposit does not include, and no verbatim participant instructions are published.
cognitive_children_2026	The item wording shipped for this table is the authors’ own English translation; the assessment was administered in Spanish in Ecuador and no Spanish wording was deposited. The 1-5 anchors are published only at the endpoints (1 = Never, 5 = Always), so options 2-4 carry no text.
coh_goodman_2023_racial_comp	Respondents chose among pictures labelled A-F, so option_text gives the article’s description of each picture; resp 8 on the four school items is undocumented, and which of each pair (school vs classroom, neighborhood vs block) is which rests on questionnaire order rather than a codebook.
colomer_perez_2021_soc13	Item text is the canonical English SOC-13 (Antonovsky Orientation to Life Questionnaire) wording; the study administered a Spanish version whose wording is not reproduced in the paper or its data deposit, and the deposited .sav carries no labels tying its INVSOC1-SOC13 codes to individual items, so the mapping rests on the code numbering plus a scoring check that pins each item’s polarity but not its position within a polarity class. Responses are stored raw, not reverse-scored, so for items 1, 2, 3, 7 and 10 a high value means a lower sense of coherence. Only the two extreme scale points are labelled on the instrument; points 2-6 are intentionally blank.
conner_2017_bfi	IRW does not offer item text for this table. In the response data, sixteen of the 44 items (BFI 2, 6, 8, 9, 12, 18, 21, 23, 24, 27, 31, 34, 35, 37, 41, 43) are stored already reverse-scored, so for those items a low resp indicates agreement with the item as worded
conner_2017_cesd	Items 4, 8, 12 and 16 are stored already reverse scored, so their response anchors are shipped in reverse order (resp 0 = ‘Most or all of the time’) relative to how the positively-worded stem reads; this matches the numbers in the response data.
conner_2017_curiosity	Item wording is transcribed verbatim from the study’s own SPSS file, so it preserves that file’s wording rather than the published Curiosity and Exploration Inventory-II: item 1 reads “I actively seeks as much information as I can in a new situation” where the published CEI-II reads “I actively seek … in new situations”, and items 1 and 8 carry a trailing full stop the other eight lack. The source file also labels the block “Curiousity and Engagement Inventory II”; the instrument field ships the correct name (Kashdan et al. 2009), which the paper’s own reference 47 confirms.
conner_2017_lot	Items 2, 4 and 5 (the pessimism-worded items) are stored already reverse-scored, so for them resp 1 = ‘Strongly agree’ and 7 = ‘Strongly disagree’, the reverse of the printed anchors; the shipped option text is flipped to match the data.
conner_2017_vitality	The IRW dictionary describes this table as the Subjective Vitality Scale; the items are in fact the modified 4-item energy/fatigue (Vitality) subscale of the RAND SF-36, and items 3 and 4 (“worn out”, “tired”) are stored already reverse-scored, so a response of 100 on those items means “None of the time”.
conspiracy_asd__cognitive_flexibility	Item wording here comes from the deposit’s plain-text survey document, not from the coded Qualtrics workbook that accompanies it. That workbook’s label row is corrupted by a global find/replace of ‘no’ to ‘2’ – it reads ‘I feel I have 2 power’ and ‘I just don’t k2w what to do’ – so anyone comparing this table against those labels will find differences. The plain-text document agrees word for word with the uncorrupted labels in the deposit’s raw data file.
conspiracy_asd__conspiracy_gcbs	The source Excel’s question-text header carries a mechanical corruption in which the letters ‘no’ were replaced by the digit ‘2’ (e.g. ‘in2cent’, ‘tech2logy’); the item text published here reverses that substitution.
conspiracy_asd__schizotypy	The item codes are a 37-item sequence, not MSS-B item numbers: the study administered all 38 MSS-B items but its deposited data omit “I believe that dreams have magical properties” (survey item 11), so schizotypy_11 onward are offset by one from the published MSS-B numbering. The transcribed instructions also say to answer “from 1 to 4” while the anchor table beneath them and the data are 1-5.
conspiracy_asd__thinking_styles	The instrument is the 10-item Rational-Experiential Inventory short form, administered under the survey’s own heading ‘Thinking Styles’. Only the two extreme scale points are labelled on the instrument (‘completely false’ and ‘completely true’), so option_text is blank for responses 2, 3 and 4 – those points are genuinely unlabelled rather than missing. Wording is taken from the deposit’s plain-text survey document, because the coded workbook’s labels carry a find/replace artifact (‘k2w’ for ‘know’).
content_literacy_intervention_g1	Only the 20 reading-motivation items (codes 1-20) carry text; the 96 science and social-studies vocabulary items (codes 21-116) are blank because no accessible source names their target words. The motivation wording is the Me and My Reading Profile adaptation printed in Appendix G of the project’s preceding first-grade trial (ERIC ED662409), since this study’s own copy of the form was not accessible. The printed form flips option direction between items, but the deposited file stores one uniform direction (3 = most positive), so for items 1, 4, 5, 7, 8, 10, 15, 17, 18 and 20 the option text is shipped in the reverse of the printed order to match the data.
cooper_2018_funny_topics	This battery is a select-all-that-apply checklist, so resp is a checkbox state rather than a chosen response option, and the option_text values ‘Selected’ and ‘Not selected’ describe that 0/1 coding rather than transcribing anything the survey printed – the source publishes no response labels for this battery, only the stem ‘Please select all that you might find funny.’ Item text and instructions are verbatim from the study’s own survey file.
cooper_2018_offensive_topics	This battery is a select-all-that-apply checklist, so resp is a checkbox state rather than a chosen response option, and the option_text values ‘Selected’ and ‘Not selected’ describe that 0/1 coding rather than transcribing survey wording – the source prints no response labels for it. Item text is verbatim from the study’s own survey file.
cordova2019_clinical_edu_environment	The item text is the administered Spanish, taken verbatim from the question headers of the study’s own response file in the Harvard Dataverse deposit – what the 391 respondents answered, and what reproduces all 200 item x resp cells. It differs on seven items from the Spanish printed in the paper’s Table 1, which is a later rendering by the authors. The English in the _translated columns is a machine translation produced by IRW, not the study’s wording, and has not been reviewed by a Spanish speaker; treat it as a reading aid. option_text is blank because the administered anchor labels were never published: the form stored 1-5 and the paper states only that the scale runs from “totally in disagreement” to “totally agree”. The instrument is the PHEEM, not the DREEM the data dictionary’s description names.
cormier_2024_personality	IRW does not offer item text for this table: the wording was the BFI-2-XS, withdrawn under the 2026-09-10 ruling that the BFI-2’s non-commercial-only terms block it. In the response data, item Q20_8 is not a personality item but an attention check instructing respondents to select ‘Agree Strongly’, and every respondent did (all 898 responses are 5). Exclude it before scoring the BFI-2-XS.
cormier_2024_phq4	The source Qualtrics export concatenates a shared block stem with each item stem in one header, joined by ’ - ’. That shared stem is shipped once in instructions and item_text holds only the per-item sentence, so an item here is shorter than the corresponding column header in the source file.
coroiu_2018_phq9	The item text is the canonical English PHQ-9. The study administered the German version to a German general-population sample, and neither the data file, the paper nor its supplements publishes the German item wording.
coroiu_2018_scs	The study administered the German SCS (SCS-D). The German item wording was not published, so the item text is the study’s own English labels (Neff’s original 2003 English items). The response options are the German anchors respondents saw (‘sehr selten’ .. ‘sehr oft’), with the study’s English (‘very rarely’ .. ‘very often’) alongside; these differ from Neff’s canonical ‘Almost never’ .. ‘Almost always’. Responses are stored raw: the paper says the negative subscales were reverse coded for its analysis, but the stored data are not.
corti_2023_academic_adaptation	The 1-5 response anchors are not shipped: neither the study’s SPSS deposit nor the paper publishes the labels for this five-point scale, so option_text is blank and the direction of the scale is not documented by the source.
covid_travel_mackenzie_2021	The resp values 0-5 carry two different scales in this table: 48 behaviour-frequency items run Never=0 to Everyday=5, while 26 attitude items run Strongly disagree=0 to Strongly agree=5, so resp should be read against each item’s own option_text rather than pooled across items
cox_2024_feedback_perceptions	Only three response anchors are shipped for what the study describes as a five-point scale: no respondent ever used 1 (never) or 2 (once in a while), so those values do not occur in the table and only the sometimes/often/almost always anchors appear. Item text is the source spreadsheet’s column headers verbatim, minus their PreQ#/PostQ# wave prefixes – one item code spans both waves – with three run-together words separated (‘feedbackin’, ‘throughoutyour’, ‘memberin’).
CPDMMC_Kunnari_2020_HCD	The codebook this table follows keys VT to the Vaccine dilemma and V to the Vitamins dilemma, which reads backwards at a glance; the assignment is the study’s own and is transcribed as published. Only the endpoint anchors are labelled (1 = Not acceptable at all, 7 = Totally acceptable) because the deposit labels no others, and instructions is blank because no literal instruction text is published anywhere in the deposit.
CPDMMC_Kunnari_2020_PDP	Two wordings that look like transcription errors are the source’s and were kept verbatim: the CA_C/CA_IC vignettes read ‘too fast to break in time’ (for ‘brake’), and VP_C reads ‘sideeffects’. Each item carries its own section_prompt rather than sharing one, because the congruent and incongruent versions of a dilemma are different scenarios; where the source vignette repeated its judgment question inside the paragraph, that duplicate sentence was dropped so the question appears once, in item_text.
CQTMS_Hur_2023	Item text covers only 79 of the 160 items – the study published wording for the finally selected items only, so the other 81 ship blank item_text; of the 79, 30 have Korean wording with no English translation because the article’s English supplement omits three of the eight character qualities. The 5-point response anchors are shipped in English (the Korean anchors were never published) and describe the raw scale direction stored in the table, which is the opposite of the reverse-scored distributions the article’s Supplement 1 reports for 17 negatively-worded items.
criticalperiod_syntax	Six of the 95 items (q1, q2, q3, q5, q6 and q7) are sentence-to-picture matching items whose two answer choices are pictures. The source publishes those pictures only as figures, never as text, so the option_text on those twelve rows is a machine-generated description of each picture rather than source wording – every one is prefixed [machine-generated image description] to make that unmistakable, and it is the only text in this table not transcribed from the source. Which picture belongs to which response is not guessed: the study’s supplement prints an answer key naming the correct panel for each of the six, and because these are two-alternative forced choices, resp = 1 means the respondent chose that panel and resp = 0 means they chose the other. The remaining 89 items are grammaticality judgments and carry no option_text at all – there resp is scored accuracy on a “choose all that apply” question, so no single option was chosen and any label would be invented. Separately, Question 10’s four response options are printed in the supplement under a duplicated “c.” label (a, b, c, c, with no “d.”), so item q10_4 is identified as the fourth printed option, “What old are you?”, by its position and by the answer key’s grammaticality value; “How old are you?” (option 3) is grammatical in every dialect and is not in the table
CSE_Bayazit_2022	The item wording is the administered Turkish, with the authors’ English alongside. The response anchors are published only in English and only at the endpoints (1 = strongly disagree, 5 = strongly agree), so the option text is English, options 2-4 carry no text, and the Turkish anchors respondents saw are not available. The table includes Item11, which the paper dropped from its final 14-item scale.
cucchi_2018_kims	Item codes KIMS1-18 are this study’s own renumbering of the 39-item KIMS Describing (1-8) and Acting with Awareness (9-18) subscales; they are not canonical KIMS item numbers.
cucchi_2018_rfq	The source file’s response anchors for RFQ1 differ from the other seven items in casing and read ‘somehow agree’ where the rest read ‘Agree somewhat’; the shipped anchors use the wording the other seven items carry.
cucchi_2018_scoff	Items SCOFF2, SCOFF3 and SCOFF4 ship without a terminal question mark and SCOFF1 and SCOFF5 with one, because that is how the source data file’s variable labels write them – not an omission to be tidied. instructions and section_prompt are blank because neither the paper nor the deposit publishes any administration instructions for the SCOFF.
cugmas_2021_elderly_social_support	resp here is a count of network members, not a rated response: each item is one relationship type and the value is how many people of that type the respondent named across the survey’s three social-support name generators, so there are no response-option labels. Note also that the study’s English-language questionnaire (S2) lists 14 relationship categories while the Slovenian questionnaire actually administered (S1) lists the seven that appear in the data; the seven Slovenian categories are what is published here.
CV_OASIS_ODSIS_PPE_Novak_2020_PANAS	Item text for this Czech-administered PANAS table is the canonical English Watson, Clark & Tellegen (1988) adjectives and response anchors, not the Czech wording respondents read, which the study does not publish; the adjectives are assigned by the canonical item order, which the data confirm to the level of positive/negative subscale (with ‘alert’ behaving like a negative-affect item in this sample, as the authors note) but which nothing in the source confirms item by item within a subscale.
CV_OASIS_ODSIS_PPE_Novak_2020_RSES	Item text for this Czech-administered Rosenberg Self-Esteem Scale table is the canonical English RSES wording, not the Czech wording respondents actually read – the OSF deposit and the paper publish neither the Czech items nor any per-item labels. The item codes are reconstructed as the widely circulated RSES ordering (reverse-keyed at 2, 5, 6, 8, 9), not Rosenberg’s 1965 ordering, on the basis of the study’s own scoring script; that pins each item’s polarity class, the response direction and RSES_8, but does not distinguish RSES_3 from RSES_4, RSES_1 from RSES_10, or the four remaining reverse-keyed items from one another. Responses are stored raw, with 1 = ‘Strongly agree’ and 4 = ‘Strongly disagree’ for every item including the reverse-worded ones.
dahlstrom_2022_scoare	Six items (SF, SD, ST, SM, KE, KA) were asked only in the workshop’s first program year, whose survey version PLOS does not publish; they carry the second-year form’s parent stem without its ‘scientific writing’ / ‘speaking or presenting’ qualifier.
daiku_2021_dirty_dozen	The DTDD-J was administered in Japanese; item_text_translated reproduces Jonason and Webster’s (2010) original English for the correspondingly numbered items, but the English shown for the instructions and for the two scale anchors was translated by IRW rather than by the study’s authors.
dalichaouche_2026_covid_attitudes	In the response data every unanswered attitude item is stored as 0 rather than missing (97 cells; 15/19/27/15/21 for A1-A5), and A1’s 0/1 codes do not follow the questionnaire’s key – 0 holds 72 respondents who answered ‘Pas du tout inquiet’ (not worried at all, keyed 1) and 46 who answered ‘Oui, je panique’ (keyed 0), while all 19 at 1 answered ‘Oui, je panique’ – so A1’s resp 0 and 1 carry no option_text; item stems are the study’s English rendering because the French stems survive only as truncated column headers.
dalichaouche_2026_covid_knowledge	Six COVID-19 knowledge questions scored 1 = correct, 0 = incorrect; the study was run in French and the English shown is the author’s own translation from the deposit, blank answers are scored as incorrect (30 of 1,800 responses), and on item C4 the data also count answers including fatigue as correct although the published key names only fever and cough.
dalichaouche_2026_covid_practices	In the response data every unanswered practice item is stored as 0 rather than missing (55 of 1,800 responses; 10/18/16/4/4/3 for P1-P6); item stems are the study’s English rendering because the French stems survive only as de-spaced column headers, while the response options are the administered French.
dalky_2020_sf36	This SF-36 was administered in Arabic; the item and option wording shipped here is the study’s own English rendering, taken from its SPSS variable and value labels, not the Arabic wording respondents read (the deposit publishes no Arabic text). The five block stems are the canonical public-domain RAND-36 English wording, supplied because the data file carries no block headers and many item labels are fragments without them. The table holds 35 of the SF-36’s 36 items – one physical-functioning item and one general-health-perception item are absent from the source file.
dasilva_2019_hexaco24	The English in the _translated columns is the original Brief HEXACO Inventory wording as reproduced by the study’s own paper, not a translation of the Portuguese actually administered; six items (3, 7, 9, 14, 15, 19) and the two ‘slightly’ anchors differ in wording between the two versions.
dass_Thiyagarajan2022	Most of this sample did not read the wording shipped here: the study let participants choose the Bahasa Malaysia or English DASS-21, and 90% of the 971 respondents used the Malay version, but only the English wording appears in the study’s deposit or paper.
dass21_medical_graduates_bangladesh	Respondents could answer either the English or the Bengali (Bangla) DASS-21, and the data do not record which; the item text here is the English version.
dassean_2025_bsmas	The English item and instruction translations (*_translated) were written by IRW from the Arabic questionnaire respondents read; the English option labels are the study’s own.
dd_rotation	The items are pictures, not text: each is a display of two 3-D objects and the respondent judges same/different, so item_text carries only the stimulus’s rotation angle (50, 100 or 150 degrees) as documented in the diffIRT package, and the option labels describe scored accuracy rather than the yes/no button the respondent pressed.
de_vries_2022_bat_burnout_core	The Dutch wording shipped here is the official work-related Burnout Assessment Tool questionnaire, not text published by this study: the study’s own deposit carries no variable or value labels for any BAT column and the paper prints only two example items. Subscale membership is exact – each subscale’s item row-means reproduce the deposit’s own BAT scores – but the position of an item within its 5- or 8-item block rests on the order the official questionnaire presents them in, and nothing in the data distinguishes one within-block position from another.
de_vries_2022_bat_secondary	As with the core BAT table, the wording is the official Dutch BAT questionnaire’s rather than the study’s: neither the deposit (no labels on any BAT column) nor the paper, which describes only the 23 core items, publishes the secondary-symptom wording. The instructions and the 1-5 anchors are likewise the canonical form’s – the scale was administered inside a larger online survey whose framing text was never published – and the assignment of batspan01-05/batsoma01-05 to positions 1-5 within each block follows the official form’s order.
de_vries_2022_eudaimonic_wellb	The study was run on Dutch employees and presumably administered a Dutch EWWS, but that wording was not published, so the item text is the original English EWWS (Bartels, Peterson & Reina, 2019). Dimension membership is confirmed by the data, but the order of items within each dimension is inferred from the item-code numbering.
debacker_2018_decisionjustification	The paper’s Measures section states items were answered on a 5-point scale from 1 (strongly disagree) to 5 (strongly agree), but the deposited data are coded 0-5 and all six levels are used (e.g. item 1: 6/28/69/123/267/175), so the published anchor labels cannot be attached to specific resp values; option_text is left blank.
deception_professors	Item 5 (coded ‘not relevant to rigor’) holds only the value 1 – respondents not given that code are missing, not 0; items 20 and 21 each keep only the first-listed code for respondents whose free-text answer received several (23 of 267 and 46 of 306); and item 17 (percentage of own designs using deception) records 0 for the 357 respondents who were never asked it because they do not use deception.
degirolamo_2022_cmq	Item and response wording is the Italian administered in the survey; the English item translations are from the study’s own English source questionnaire (its final word for item pag14_1_5 restored from the published CMQ), while the English instructions and response labels were translated by IRW.
degirolamo_2022_tipi	Item and response wording is the Italian revised Ten-Item Personality Inventory (I-TIPI-R) as administered; the instructions are cut off mid-word because the source codebook truncates them, English for items 5, 8 and 10 (revised in the Italian form), the instructions and the response labels was translated by IRW, and the other seven English items are Gosling et al.’s original wording.
deilkas_2019_collaboration	As for deilkas_2019_saq: the deposit’s English labels, not the Norwegian wording respondents read.
deilkas_2019_patient_safety_climate	This Safety Attitudes Questionnaire was administered in Norwegian, but no Norwegian wording appears anywhere in the study’s deposit or supplements, so the item and option text shipped here is the study’s own English, taken from the SPSS variable and value labels of its deposit – not the wording respondents read. Separately, Q34 and Q59 ship byte-identical item_text: their source variable labels differ only by the questionnaire numbering prefix they carry (‘31.’ vs ‘56.’), which is stripped on the way in. Q59/Q60 are the instrument’s usual emergency/routine fatigue pair, so it is Q34’s label that is most likely a copy-paste error in the deposit; it is shipped as labelled rather than blanked, and its item text should be treated as uncertain.
deilkas_2019_saq	The item text is the English wording in the deposit’s SPSS labels; the SAQ-Nursing Home was given in Norwegian, and the Norwegian wording respondents read is not published.
dejesus_2017_lequesne	The study deposited both the Portuguese questionnaire it administered and its own English rendering, and the two disagree at three points: the everyday-life squat item reads ‘Para agachar-se completamente’ in Portuguese and ‘Can you pick up an object from the floor?’ in English, the morning-stiffness anchors differ, and the walking-distance anchor for level 1 reads ‘superior a 1 hora’ against ‘> 1 kilometer but limited’. Both are shipped as written, with the administered Portuguese in item_text and the study’s English in item_text_translated. The maximum-distance item is scored 0-6 by distance band with 1 or 2 added for walking aids, so its levels 7 and 8 are composites with no published label and ship blank.
dejesus_2017_sf36	The Brazilian-Portuguese questionnaire this study administered prints 1 = Sim / 2 = Nao for the yes/no items of questions 4 and 5, but the deposited data stores them as 0/1, so the option codes here follow the data rather than the printed form. That direction was established from the data, not assumed: for all seven yes/no items the resp=1 group has the worse physical-functioning block mean (1.49-1.61 against 2.16-2.29). The table also holds only 22 of the SF-36’s 36 items – questions 1-8. Questions 9-11 were administered, since the paper reports domain scores that need them, but they are absent from the study’s deposited data file.
depression_anxiety_stress	This table pools three instruments administered in one openpsychometrics session, not the DASS alone: items 1-42 are the DASS-42, items 43-52 are the Ten Item Personality Inventory, and items 53-68 are a 16-word vocabulary checklist scored 0/1 (VCL6 “cuivocal”, VCL9 “florted” and VCL12 “verdid” are not real words and were included as a validity check). The item codes are bare integers assigned by the processing script and carry no content of their own
depue_2023_gds15	Respondents answered the Dutch translation of the GDS-15 (Bleeker et al.), which the study does not publish; the item text is Yesavage’s original English standing in for it. Responses are stored key-scored, 1 = the depressive answer, so for items GDS1, GDS5, GDS7, GDS11 and GDS13 a 1 means the respondent answered NO, and for the other ten a 1 means YES.
depue_2023_pwi	Respondents answered the Dutch version of the PWI-A, which the study does not publish; the item text is the International Wellbeing Group’s English standing in for it. Items PWI1-PWI8 were rated for the past month (at each of three waves), while PWI1_pre-PWI8_pre are the same questions rated retrospectively, at the first wave, for the period before COVID-19; the study’s own wording for those two reference periods is not published, so both sets carry the same stem.
depue_2023_subjcog	Respondents answered Dutch questions that the study does not publish; the item text is the study’s own English description from its data readme, not necessarily a literal translation. The five ‘problems with …’ items refer to ‘the COVID-19 period’ at the first wave and to ‘the past month’ at waves 2 and 3, and CognFunct_now’s wave-3 wording omits ‘(so DURING the COVID-19 lockdown)’. The 0-10 rating runs from 0 = ‘very bad’ to 10 = ‘very good’; no respondent chose 0, so only the 10 anchor has an option row.
derubeis_2017_arsq	The wording respondents read is the German in item_text, taken verbatim from the deposited SPSS file’s own variable labels; the English in the _translated columns was generated by IRW, not by the study, and has not been checked against a published German-to-English ARSQ. Beyond that, the ARSQ presents 20 hypothetical situations, each rated on two sub-items (A = how anxious you would be, B = how likely the other person is to respond positively), and the situation vignettes themselves appear neither in the study’s deposit nor in the paper – so section_prompt is blank and only the two sub-item questions are given. Scale points 2-5 are unlabelled in the source and ship blank.
di_riso_2025_contact_behavior	Administered in Italian, but only the authors’ English rendering was published (read from an image-only table in the paper), so item text is English. The three statements are separate attitude items, not a unidimensional scale: Contact_with_professionals is approving and correlates negatively with the other two (-0.13 and -0.07). Scale points 2-4 are unlabelled in the source.
di_riso_2025_mask_emotion	The instrument is an ad hoc block administered in Italian, and no Italian wording was published; the shipped item text is the English one-word emotion label the paper’s Table 2 uses (Controlled, Weak, Scared, Silly, Brave, Caring, Strong, Protected), not a full administered item stem, and scale points 2-4 are unlabelled in the source.
ding_2025_iu	The scale was administered in Chinese (C-IUS-12), which is not published; the English shown is the study’s own for 11 items. IU9’s wording is not printed by the study: it is the standard IUS-12 item ‘When it’s time to act, uncertainty paralyses me’, inferred by elimination.
dinic_2025_shortdarktriad	This table pools 14 samples administered in 12 languages, so the item text shown is the English SD3 master wording rather than the strings most respondents read, and no single administered language could be named. At items sd3_2, sd3_7, sd3_8 and sd3_26 the study’s own two sources disagree: the paper’s Table 3 prints the canonical SD3 wording, which is what ships, while the deposited SPSS file’s variable label at those four positions quotes a statement that is not an SD3 item at all (for sd3_26, ‘I like to pick on losers.’). The other 23 are identical in both. Responses for sd3_11, sd3_15, sd3_17, sd3_20 and sd3_25 are stored already reverse-recoded, so for those five resp 1 means ‘Agree strongly’ and 5 ‘Disagree strongly’ – the opposite of the other 22.
direkvand_2022_mjsi	This instrument was administered in Persian and the Persian version is deposited, but it could not be attached to the item codes: it numbers its items 1-25 continuously and groups the professional and physical-mental items differently from the paper’s Table 4, which is the numbering the codes follow, and three items differ outright between the two. So the study’s own English version ships instead of the wording respondents read. No response-option text is shipped at all: the published instrument prints one uniform 5-point scale for every item, while the deposited SPSS file’s value labels give most items a 2-, 3- or 4-category scale that matches the response ranges actually observed – the two accounts contradict each other and neither could be established, so resp 1-5 ships unlabelled. Item-to-code matching is verified at the level of the five dimensions and not for the order of items within a dimension.
disgust_berger2014	The Disgust Scale-Revised as administered in Hebrew. Two sections: items 1-14 are agreement statements and 15-27 are ratings of how disgusting an experience would be, both 0-4 but with different anchor wording. Items 12 and 16 are the scale’s catch items and are not in the data. Items Q1, Q6 and Q10 are reverse-keyed and are stored UN-reversed here, so a high value on those three means low disgust sensitivity; the standard scoring subtracts them from 4. The English alongside is the published DS-R, not a translation of this Hebrew form.
DMCT_Addis_2020_PSIQ	Item text is English while the study administered the Psi-Q in German (no German wording is published in the deposit or paper), and within each sensory modality the item order follows the Psi-Q form’s own numbering - the Spanish and Japanese validation appendices publish a different within-block order that this study’s data cannot rule out, so a residual mismatch would be confined to items within one modality.
doherty_2023_dass21	The item and option wording is the deposit’s SPSS labels, which differ from the canonical DASS-21 in three places, kept as found: the resp=1 anchor reads ‘or a good part of time’ (canonical ‘or some of the time’), making it nearly identical to the resp=2 anchor; DASS4 omits ‘breathlessness’; and DASS9 reads ‘I worried’ for ‘I was worried’. Whether these are the administered wording or labelling shorthand cannot be determined. The resp=2 label is cut off in the source file and was completed from the canonical DASS anchor.
dong_2025_teacher_leadership	The 16 items are shipped in the authors’ English wording from the S1 Data column headers; the questionnaire was administered to teachers in Xinjiang, China and no Chinese wording appears in the article or either supplementary workbook, so this is an English substitute for what respondents read. The paper’s own prose quotes some of the same items in a slightly different English rendering.
dopmeijer_2022_performance_pressure	The item and response-option text is the study’s own English rendering from its SPSS file; the survey was administered in Dutch, and the Dutch wording is not published. The paper analyses only one of these twelve items (PerformancePressure_A), quoting it as ‘To what extent do you feel that you have to perform or achieve?’, and its reported mean for that item is on a 0-4 scale although it describes the scale as 1-5.
dou2025_ans	The English item text for this table was produced by IRW, not by the study authors; the administered wording is the Chinese in item_text. No response-option wording is shipped: respondents answered on a six-point scale whose anchors are published neither in the deposit nor in the paper, and the paper’s stated ‘seven-point … 1 = strongly disagree, 7 = strongly agree’ describes the original Aesthetic Needs Scale rather than the six-point form used here (the deposited data contain only the values 1-6).
dou2025_area	The English item_text_translated for this table was produced by IRW, not by the study authors; and its response anchors are shipped as the paper’s English gloss (1 = strongly disagree, 5 = strongly agree) because the Chinese wording respondents actually saw for the scale points is not published in the deposit or the paper.
doustmohammadian_2017_fnlit	Two things to know about this table’s item text: 16 of the 58 items were dropped during validation and never appear in the published Persian or English FNLIT questionnaires, so they ship English-only wording taken from the paper’s S1/S2 item-analysis tables while the other 42 carry the administered Persian with the authors’ English alongside; and Interactive_familyDebate is stored reverse-scored, so its anchors run ‘always’=0 to ‘never’=4, the opposite of every other frequency item and of the printed questionnaire.
dpt_noncog__emotional_intelligence	This table holds 14 of the 33 SSEIT items, and the codes preserve the instrument’s original item numbers rather than renumbering 1-14. The wording is the source workbook’s own header row transcribed verbatim, including two places where it departs from canonical SSEIT wording – schutte_2 reads ‘I remember times I faces similar obstacles’ (canonically ‘I faced’) and schutte_13 reads ‘I arrange events other enjoy.’ (canonically ‘others enjoy’) – neither corrected here. Only the two endpoints of the 1-5 scale were ever labelled by the source (1 = strongly disagree, 5 = strongly agree), so option_text is blank for 2-4.
dpt_noncog__intolerance_of_uncertainty	The deposit labels two different items ‘IU_Scale_21’, so IRW’s codes iu_21a and iu_21b are a disambiguation, not IUS-27 item numbers – iu_21b (‘Being uncertain means that I lack confidence.’) is item 22 in the published IUS-27, and item codes here should not be read as instrument item numbers.
dss_mouta_2021	The item text is the administered Brazilian Portuguese form from the study’s OSF deposit. The English in the _translated columns has two sources. The items are the original Hamilton, Shih & Mohammed (2016) English DSS, a parallel version rather than a translation of the Portuguese, so the phrasing differs in places (item 2 ‘instinto’ vs ‘gut feelings’). The English instructions and the three middle response labels were machine-translated by IRW.
dudasova_2021_cpc12	Study 1 administered the original CPC-12 in Czech, but only the revised CPC-12R Czech wording was published; the three original resilience items (PsyCap7-9) are therefore shown in canonical English rather than the Czech respondents actually read.
dudasova_2021_cpc12_study3	The table name’s ’_study3’ refers to the article’s S3 supporting-information file, not to the paper’s Study 3: these are the 202 German respondents analysed in Study 2, who completed the original German CPC-12, not the revised Czech CPC-12R. The 6-point response anchors are shipped in English because only the English endpoint labels (‘strongly disagree’ / ‘strongly agree’) were ever published; the administered German anchor wording is not in the paper or in any deposited file.
dudasova_2021_gratitude	The source study never names or reports this gratitude measure and its data file carries no item labels, so the instrument was identified structurally as the GQ-6 and the wording shipped here is the canonical English GQ-6, not a string any respondent read – the survey was administered in Czech and no Czech version is deposited. Note also that the stored values for Grat3i are already reverse-scored while Grat6 is stored raw, so Grat3i’s response anchors run in the opposite direction to the other five items.
dulger2024_wordcompletion	For 36 of the 60 word fragments the target word in correct_response was derived by IRW from the modal correct completion in the study’s own coded data rather than stated by the authors; it agrees with all 24 fragments whose key the study does state, and each derived word reproduces its item’s observed accuracy (e.g. ac_e_ted 0.943 derived vs 0.950 live).
dumas_Organisciak_2022	The paper describes originality as coded on a 0-4 scale, but the IRW table (and option_text) is on 1-5. This is not a shift introduced here: the authors’ own deposited file (osf.io/qcmex) already codes rater1-rater4 as 1-5, and the processing script does no recoding, so the table reproduces the source faithfully. The discrepancy is between the paper’s text and the authors’ shared data; read the 1-5 values as the paper’s 0-4 categories shifted by one
dussel_2022_pcl17	This PCL was administered in French; the item text, instructions and anchors shipped here are the authors’ own English rendering of the questionnaire (S1 Appendix), not the French wording respondents read, which the deposit does not publish. Note also that the S1 Appendix numbers the response scale 0-4 while the response data are stored 1-5 (the standard PCL coding, confirmed by the 17-85 total range and by 13.7% of respondents exceeding the paper’s PCL>44 cutoff against the 14% reported); the anchors are shipped against 1-5. Finally, the two hyperarousal item codes PCL_defensive and PCL_watchful are both plausible names for item 16 (‘Being “super-alert” or watchful or on guard’); they are assigned to items 16 and 17 by their position in the source file, and no check in the data distinguishes them.
dvivdtws_ppmial_marcatto_2023_dtw	This table pools two administrations of the Dark Tetrad at Work scale, an Italian-language sample (cov_language = 1, n = 300) and an English-language sample (cov_language = 2, n = 253), but item_text is the Italian wording for every row. The English-speaking respondents read the original English DTW wording, which is in item_text_translated. The Italian response anchors and instructions were never published, so option_text and instructions are blank; the paper states the scale ran 1 = strongly disagree to 5 = strongly agree.
dwyer_2019_clinton_activist	Respondents answered the Activist Identity and Commitment Scale on a 0-to-5 scale, but IRW stores the SPSS codes 1-6, so each resp value is one higher than the number on the scale the respondent saw; option_text gives the anchor wording for the corresponding point.
eammi_grahe_2018_marriage_identity_allocation	Respondents typed a percentage for each of the four roles, so there are no response labels. The allocation was meant to total 100, but the data do not enforce it: in the source workbook 207 of 3,178 respondents’ four entries sum to more than 100, and in the live table about 9% of respondents do not sum to 100 (maximum 400). The IRW processing script dropped 4 entries above 100, which leaves 13 respondents with only three of the four items.
eammi_grahe_2018_physsx	The response table is missing the top response category. Respondents answered on a 3-point scale (1 = not bothered at all, 3 = bothered a lot), but the IRW processing script dropped all 6,464 responses of 3 (15.6% of all answers, and 1,552 of 3,177 on physSx_12), so the table holds only 1 and 2, and 2 is the unlabelled middle point, not ‘bothered a lot’. The 13 items are the PHQ-15 somatic symptoms minus the two sex-specific ones.
ecps_sahm_2024_distrust	Item text and the 7-point agreement ladder (Strongly disagree, Disagree, Slightly disagree, Neutral, Slightly agree, Agree, Strongly agree) are the administered English wording. The 6 misperception items are matched by the form’s own B.x.y numbering; the 4 conspiratorial-thinking and 3 anti-expert items by listing order. Note on the table name: the ecps_sahm_2024_ prefix points at Sahm et al.’s German Emotion Regulation Questionnaire validation, which is not what this table is. These responses are the COVIDiSTRESS Global Survey Round II; its cleaned data file is hosted inside that study’s OSF project, which is how the two came to be conflated. The name is kept because renaming a published table breaks every existing reference to it.
ecps_sahm_2024_dtl	Item text and both ladders are the administered English wording. The table pools two different scales: the 6 resilience items run 1-7 on an agreement ladder, and the 5 uncertainty items run 1-5 on a ‘characteristic of me’ ladder, so resp is not comparable across the two blocks – read section_id and option_text rather than assuming one scale. Note on the table name: the ecps_sahm_2024_ prefix points at Sahm et al.’s German Emotion Regulation Questionnaire validation, which is not what this table is. These responses are the COVIDiSTRESS Global Survey Round II; its cleaned data file is hosted inside that study’s OSF project, which is how the two came to be conflated. The name is kept because renaming a published table breaks every existing reference to it.
ecps_sahm_2024_emotion	The Emotion Regulation Questionnaire short form, 8 items: expressive suppression (items 1, 2, 4 and 5) and cognitive reappraisal (3, 6, 7 and 8). resp 0-6 is a 7-point agreement scale whose midpoint the cleaned data recodes to 0, so 0 = Neutral, 1 = Strongly disagree and 6 = Strongly agree; the form labelled only those three positions, so the four between them carry no option_text. Unlike its ten sibling tables this one really is the Emotion Regulation Questionnaire the dataset description names.
ecps_sahm_2024_ia	Item text and the response ladder are the administered English wording of the COVIDiSTRESS Round II form. resp runs 1-10 on a slider whose ends only are labelled, 1 = ‘Not at all’ and 10 = ‘Frequently’; the eight intermediate positions had no verbal label and so carry none here. Items are tied to codes by the form’s listing order.
ecps_sahm_2024_identity	Item text is the administered English wording. resp 0-6 is a 7-point agreement scale whose midpoint the cleaned data recodes to 0: 0 = Neutral, then 1 = Strongly disagree through 6 = Strongly agree. Note that 0 is therefore the MIDDLE of the scale, not its bottom. Note on the table name: the ecps_sahm_2024_ prefix points at Sahm et al.’s German Emotion Regulation Questionnaire validation, which is not what this table is. These responses are the COVIDiSTRESS Global Survey Round II; its cleaned data file is hosted inside that study’s OSF project, which is how the two came to be conflated. The name is kept because renaming a published table breaks every existing reference to it.
ecps_sahm_2024_moral	Item text is the administered English wording. resp 0-6 is a 7-point agreement scale whose midpoint the cleaned data recodes to 0: 0 = Neutral, then 1 = Strongly disagree through 6 = Strongly agree. Note that 0 is the MIDDLE of the scale, not its bottom. Note on the table name: the ecps_sahm_2024_ prefix points at Sahm et al.’s German Emotion Regulation Questionnaire validation, which is not what this table is. These responses are the COVIDiSTRESS Global Survey Round II; its cleaned data file is hosted inside that study’s OSF project, which is how the two came to be conflated. The name is kept because renaming a published table breaks every existing reference to it.
ecps_sahm_2024_sscd	Item text and ladders are the administered English wording; the compliance items are past tense as administered. item_family separates compliance (1) from injunctive norms (2) and descriptive norms (3), and the three blocks use different ladders, so resp is not comparable across them. The norm ladders label only their ends, so their intermediate positions carry no option_text. Item 7 of each block is reverse-worded (‘Met with people outside of your household for non-essential reasons’) and runs against the other seven. Note on the table name: the ecps_sahm_2024_ prefix points at Sahm et al.’s German Emotion Regulation Questionnaire validation, which is not what this table is. These responses are the COVIDiSTRESS Global Survey Round II; its cleaned data file is hosted inside that study’s OSF project, which is how the two came to be conflated. The name is kept because renaming a published table breaks every existing reference to it.
ecps_sahm_2024_stress	Item text and ladders are the administered English wording for the survey’s two COVID stressor blocks, primary and secondary, both on 0 = Not at all concerned to 4 = Very concerned. The response table also holds ten Perceived Stress Scale items (perceived_stress_sca_1 to perceived_stress_sca_10), which have no item text here. Two secondary-stressor items, indices 8 and 9, carry only 6 and 8 responses against about 9,700 for the rest of their block, which is a defect in the source’s display logic rather than in the item text. Note on the table name: the ecps_sahm_2024_ prefix points at Sahm et al.’s German Emotion Regulation Questionnaire validation, which is not what this table is. These responses are the COVIDiSTRESS Global Survey Round II; its cleaned data file is hosted inside that study’s OSF project, which is how the two came to be conflated. The name is kept because renaming a published table breaks every existing reference to it.
ecps_sahm_2024_support	Three perceived-social-support items. resp 0-6 is a 7-point agreement scale whose midpoint the cleaned data recodes to 0: 0 = Neutral, then 1 = Strongly disagree through 6 = Strongly agree. Note that 0 is the MIDDLE of the scale, not its bottom. Note on the table name: the ecps_sahm_2024_ prefix points at Sahm et al.’s German Emotion Regulation Questionnaire validation, which is not what this table is. These responses are the COVIDiSTRESS Global Survey Round II; its cleaned data file is hosted inside that study’s OSF project, which is how the two came to be conflated. The name is kept because renaming a published table breaks every existing reference to it.
ecps_sahm_2024_trust	Seven institutional-trust items. resp 0-10 is a percentage slider in ten-point steps: 0 = ‘No trust’, then 10% at resp 1 through 90% at resp 9. The form’s label for the top position is not recoverable from the deposited PDF, so resp 10 carries no option_text, though the arithmetic of the scale makes it 100%. Note on the table name: the ecps_sahm_2024_ prefix points at Sahm et al.’s German Emotion Regulation Questionnaire validation, which is not what this table is. These responses are the COVIDiSTRESS Global Survey Round II; its cleaned data file is hosted inside that study’s OSF project, which is how the two came to be conflated. The name is kept because renaming a published table breaks every existing reference to it.
ecps_sahm_2024_vaccine	Item text and ladders are the administered English wording, including the form’s typo ‘how wiling you were’ in the willingness item. The table pools two scales: willingness on resp 0-4, where 0 is the Neutral midpoint, 1 = ‘Not willing at all’ and 4 = ‘Very willing’ with two unlabelled positions between; and 6 attitude items on resp 0-6, where 0 is the Neutral midpoint and 1-6 run Strongly disagree to Strongly agree. The attitude codes are numbered 2-6 and 9, so three of the instrument’s attitude positions are absent from the data. Note on the table name: the ecps_sahm_2024_ prefix points at Sahm et al.’s German Emotion Regulation Questionnaire validation, which is not what this table is. These responses are the COVIDiSTRESS Global Survey Round II; its cleaned data file is hosted inside that study’s OSF project, which is how the two came to be conflated. The name is kept because renaming a published table breaks every existing reference to it.
ecuador_2011_safety_avoidance	The English item, instruction and response-option wording for this table was produced by IRW, not by INEC, which publishes the ENVIPI 2011 questionnaire in Spanish only; the administered Spanish is in the base fields. Note also: I50914 is the form’s write-in ‘Otro ?’ (Especifique) catch-all, so its yes responses cover whatever other activity each respondent named; No aplica, don’t-know and no-answer responses were dropped, so per-item n varies (45,794-115,698), most for activities many respondents never do (I50904, I50910, I50913).
ecuador_2011_safety_confidence	The English item, instruction and response-option wording for this table was produced by IRW, not by INEC, which publishes the ENVIPI 2011 questionnaire in Spanish only; the administered Spanish is in the base fields. Note also: item I51533 is INEC’s own column name for form option 03 (Dirección Nacional Antinarcóticos), and there is no I51503; only points 1 (Ninguna confianza) and 10 (Total confianza) of the 1-10 scale have labels on the form, so 2-9 have no option text; No aplica and no-answer responses were dropped.
ecuador_2011_safety_crime	The English item, instruction and response-option wording for this table was produced by IRW, not by INEC, which publishes the ENVIPI 2011 questionnaire in Spanish only; the administered Spanish is in the base fields. Note also: these nine items (questions 7.5-7.10) ask whether the respondent suffered each specific crime between September 2010 and August 2011, not about the circumstances or reporting of the most recent crime, as the table’s description says; five of them (fraud, threat, injury, kidnapping, other) have exactly the same yes counts as the matching crime types in ecuador_2011_safety_victimtype, so the two tables largely duplicate each other on those five.
ecuador_2011_safety_homesec	The English item, instruction and response-option wording for this table was produced by IRW, not by INEC, which publishes the ENVIPI 2011 questionnaire in Spanish only; the administered Spanish is in the base fields. Note also: I51011 is the form’s write-in ‘Otro’ (Especifique) catch-all, so its yes responses cover whatever other security measure each respondent named; No aplica, don’t-know and no-answer responses were dropped, so per-item n varies, most for car security (I51007, n = 64,223, since households without a car answered No aplica).
ecuador_2011_safety_impact	The English item, instruction and response-option wording for this table was produced by IRW, not by INEC, which publishes the ENVIPI 2011 questionnaire in Spanish only; the administered Spanish is in the base fields.
ecuador_2011_safety_operatives	The English item, instruction and response-option wording for this table was produced by IRW, not by INEC, which publishes the ENVIPI 2011 questionnaire in Spanish only; the administered Spanish is in the base fields. Note also: resp 1 = Sí, 0 = No (a recode of form code 2); don’t-know and no-answer responses were dropped.
ecuador_2011_safety_places	The English item, instruction and response-option wording for this table was produced by IRW, not by INEC, which publishes the ENVIPI 2011 questionnaire in Spanish only; the administered Spanish is in the base fields. Note also: the questionnaire labels only scale points 1 (Muy inseguro) and 5 (Muy seguro), so points 2-4 have no option text, although INEC’s data file labels them Inseguro, Ni seguro ni inseguro and Seguro; No aplica, don’t-know and no-answer responses were dropped, so n is much lower for places many respondents never go (I5303 educational centre 31,658; I5308 car 53,227; I5311 entertainment venues 48,286; I5312 shows 58,404).
ecuador_2011_safety_safety	The English item, instruction and response-option wording for this table was produced by IRW, not by INEC, which publishes the ENVIPI 2011 questionnaire in Spanish only; the administered Spanish is in the base fields. Note also: the form prints masculine anchors for the neighbourhood item (I51) and feminine anchors for the city item (I52), and each ships as printed; don’t-know and no-answer responses were dropped.
ecuador_2011_safety_surroundings	The English item, instruction and response-option wording for this table was produced by IRW, not by INEC, which publishes the ENVIPI 2011 questionnaire in Spanish only; the administered Spanish is in the base fields. Note also: I51906 is the form’s write-in ‘Otros? Especifique’ catch-all, and the write-in content is not in the table; don’t-know and no-answer responses were dropped, so per-item n varies (112,498-116,789).
ecuador_2011_safety_trend	The English item, instruction and response-option wording for this table was produced by IRW, not by INEC, which publishes the ENVIPI 2011 questionnaire in Spanish only; the administered Spanish is in the base fields. Note also: resp 1 = increased, 2 = stayed the same, 3 = decreased; don’t-know and no-answer responses were dropped.
ecuador_2011_safety_victim	The English item, instruction and response-option wording for this table was produced by IRW, not by INEC, which publishes the ENVIPI 2011 questionnaire in Spanish only; the administered Spanish is in the base fields. Note also: I71A covers 2006-2010 and I72A September 2010 to August 2011; resp 1 = Sí, 0 = No (a recode of form code 2).
ecuador_2011_safety_victimtype	The English item, instruction and response-option wording for this table was produced by IRW, not by INEC, which publishes the ENVIPI 2011 questionnaire in Spanish only; the administered Spanish is in the base fields. Note also: I7406 is the catch-all ‘Otro delito ? Especifique’, and the write-in content is not in the table; five of the six crime types have exactly the same yes counts as the matching screener items in ecuador_2011_safety_crime, so the two tables largely duplicate each other; 14 responses in INEC’s file say yes to a crime type for respondents who said no to any crime in the same period (question 7.2), reproduced as published.
EEN_Lacey_2024_Children	Two of the 13 items ship less than their administered wording: helpline_willcall is cut off mid-phrase because the source Stata label hits the 80-character cap, and Howcomfortabledidyoufeel carries the deposit internal label Comfort during call rather than the question respondents heard; the study publishes neither in full. Note also that anxiety_major, depression_major and mh_major are derived GAD-7/PHQ-8 cutoff indicators rather than questions anyone was asked, and that one source typo (‘confortable’) was corrected in the shipped anchor text where every other string is verbatim.
EEN_Lacey_2024_Parent	Item wording is drawn from the study’s own Stata variable labels (abbreviated telegraphic phrasing for 14 of 24 items) supplemented by the IZA working paper’s Appendix 3 for 10 items; anxiety_major, depression_major and mh_major are derived GAD-7/PHQ-8 cutoff indicators rather than questions respondents were asked, and the language the phone survey was actually administered in is not documented anywhere in the deposit or the paper.
ees_sarling_2024	Swedish Empathic Experience Scale, thirty items in two subscales – Vicarious Experience (VE1-VE15) and Intuitive Understanding (IU1-IU15) – administered interleaved, VE at odd positions and IU at even, rated 1 = Stammer inte alls to 5 = Stammer fullstandigt. The English alongside is the instrument’s original wording from Innamorati et al. (2019), reproduced by the study’s own supplement.
ehealth_rioux_2025_gad	GAD8 is not an eighth symptom item but the GAD-7’s unscored functional-difficulty question, and all eight items are stored 1-4 (1 = Not at all / Not difficult at all) rather than the standard 0-3 GAD-7 scoring.
ehealth_rioux_2025_phq	PHQ10 is not a tenth symptom item but the PHQ-9’s unscored functional-difficulty question, and all ten items are stored 1-4 (1 = Not at all / Not difficult at all) rather than the standard 0-3 PHQ-9 scoring.
ehealth_rioux_2025_rse	RSE1..RSE10 are the ten boxes of one select-all-that-apply question about the past year: resp = 1 means the box was ticked and resp = 0 means it was not, respondents who ticked no box at all (including ‘None of the above’) are not in the table, and the four who ticked ‘Prefer not to answer’ (RSE_noanswer) are coded 0 on every event rather than missing.
eldor_2022_anomie	The survey was administered in Norwegian (a forward-back translation of the established English scale), but the deposit and paper publish only English wording, so the item text is the study’s English variable labels rather than what respondents read. Only response options 1 (completely disagree), 4 (neither agree nor disagree) and 7 (completely agree) carry labels.
eldor_2022_collective_anger	The survey was administered in Norwegian, but the deposit and paper publish only English wording, so the item text is the study’s English variable labels (its back-translation) rather than what respondents read. Only response options 1 (completely disagree), 4 (neither agree nor disagree) and 7 (completely agree) carry labels.
eldor_2022_political_resilience	The survey was administered in Norwegian and the deposit publishes only the study’s English forward-back translation for 48 of the 49 items, so the item text shows what respondents read only for item 42, whose Norwegian wording survived in the deposit (with the English in item_text_translated); the English labels’ typos are kept as printed and differ from the copy-edited English in the paper’s Table 1. Item 49 is an attention check instructing respondents to choose ‘strongly disagree’ (233 chose 1), not a resilience item. Only response options 1 (completely disagree), 4 (neither agree nor disagree) and 7 (completely agree) carry labels.
eldor_2022_realistic_threat	The survey was administered in Norwegian, but the deposit and paper publish only English wording, so the item text is the study’s English variable labels (its back-translation) rather than what respondents read. realistic_threat_3 reads ‘My ethnic group is unsafe of other ethnic groups in Norway’ in the data file but ‘unsafe due to’ in the paper. Only response options 1 (completely disagree), 4 (neither agree nor disagree) and 7 (completely agree) carry labels.
eldor_2022_relative_deprivation	The survey was administered in Norwegian, but only the study’s English wording is published, so the item text is its English variable labels rather than what respondents read. relative_deprivation_5 has no item text: the data file labels it identically to relative_deprivation_4 (‘I think my ethnic group is disadvantaged because others oppress them.’), while the scale it was adapted from (Obaidi et al., 2019) has a different fifth item (‘…disadvantaged because the West keeps them down’), and the two columns are different responses. Only response options 1 (completely disagree), 4 (neither agree nor disagree) and 7 (completely agree) carry labels.
eldor_2022_symbolic_threat	The survey was administered in Norwegian, but the deposit and paper publish only English wording, so the item text is the study’s English variable labels (its back-translation) rather than what respondents read; the paper quotes the first item differently, as ‘My ethnic culture is threatened by other groups in Norway’. Only response options 1 (completely disagree), 4 (neither agree nor disagree) and 7 (completely agree) carry labels.
eldor_2022_violent_intentions	The survey was administered in Norwegian, but only the study’s English wording is published, so the item text is its English variable labels (double spaces kept as printed) rather than what respondents read. Items violent_beh_intentio_4, _5 and _7 are negatively worded and stored unreversed. Only response options 1 (completely disagree), 4 (neither agree nor disagree) and 7 (completely agree) carry labels.
elshafie_2020_edsc	The English translations (item_text_translated, option_text_translated) were written by IRW from the Arabic checklist used in the interviews; they are not the English Baroda Developmental Screening Test items from which that checklist was adapted.
emidy2024_fevs	The .dta this table is built from carries Stata variable labels truncated at 80 characters, so the item wording shipped here is OPM’s own published 2022 FEVS instrument text (Appendix B of the 2022 FEVS Technical Report), not the data file’s labels; each of the 93 truncated labels was checked to be a prefix of the report wording for the same q-number, and the labels were used only to confirm the code-to-item tie. The instrument’s non-substantive options (Do Not Know / No Basis to Judge / Not Applicable) are unlabelled sentinels dropped by the processing script and so are deliberately absent from option_text. Note that q12 and q34 are negatively worded and already reverse-coded in the source file, so their anchors run 1 = Strongly Agree to 5 = Strongly Disagree while the other agreement items run the other way
emobank_buechel_2017	In this table id is an EmoBank sentence, not a person, and item is the rating scale (Pleasure/Arousal/Control) crossed with the annotation perspective (writer/reader), so the item text shown is the SAM scale prompt the crowdworker read, not the sentence being rated – that stimulus text is keyed to id in the EmoBank deposit’s corpus/raw.csv.
enders_2022_conspiracy_thinking	The study’s appendix lists the four conspiracy-thinking statements in a different order from the one shipped. The data reject the appendix order, so the item codes CTSCALE_1-4 are assigned from the response data instead: the order that matches the item-agreement and correlation pattern of an independently labelled sample of the same scale (CTSCALE_1 few people run things, 2 not known to voters, 3 big events, 4 plots hatched). CTSCALE_3 vs CTSCALE_4 is the least certain pair.
enders_2022_covid_conspiracy	COVCONS_1 is a seventh COVID-19 conspiracy item that the survey asked but the article never prints (it is excluded from the published 6-item scale), so its item text is blank. The article’s Table 1 numbers the other six by popularity, not by item code (its ‘Item 1’ is COVCONS_3). No instructions are published for this block, and only ‘agree’ (4) and ‘strongly agree’ (5) have option labels.
enders_2022_covid_vax_misinfo	COVVAXINFO_1 and COVVAXINFO_3 are the infertility and vaccine-shedding items, in an order the source does not reveal (both are 11% agree in the article), so their item text is blank rather than guessed. The article’s Table 1 numbers items by popularity, not by item code, and prints 12% for the DNA item (COVVAXINFO_5), where the data give 12.6%. Only ‘agree’ (4) and ‘strongly agree’ (5) have option labels.
enders_2022_science_literacy	The survey stored each respondent’s true/false answer, not whether it was correct: resp 1 means the respondent said ‘True’, and correct_response gives the published key. The appendix’s ‘(1=correct; 0=incorrect)’ describes the authors’ scored variable, not the stored response. Items are matched to codes by the appendix’s order, which the data confirm for true- vs false-keyed statements but cannot check within each group.
enders_2022_vaccine_hesitancy	The study’s appendix prints the response scale as 1 = strongly agree, 5 = strongly disagree, but the stored responses run the other way (1 = strongly disagree, 5 = strongly agree), and the option text follows the data. Items 5, 9 and 10 are reverse-worded and stored unreversed.
enem_2013_1mil_ch	The wording is INEP’s standard (AZUL) booklet, as candidates read it; 2013 has no accessibility edition. Five items (25217, 44785, 51365, 51415 and 51417) carry a description of their stimulus that was generated for this project rather than published by INEP, and each is marked inline ‘(gerada por IA)’ so it can be separated from INEP’s own words. Their stimulus is a photograph, painting or map that the printed booklet never describes, so the item previously held only the credit line and the question and could not be answered from text. Three of the five are mixed: where the picture contains printed words – a legend, a caption, a panel’s wording – those words are transcribed and sit outside the marked paragraph, because reading words off the page is recovery rather than generation. Every description was checked against the booklet twice by independent reviewers; 51365 is medium confidence because its two embedded images are only 396x243 and 237x263 pixels, the limit of what INEP’s file contains. No English ships.
enem_2013_1mil_cn	The wording is INEP’s standard (AZUL) booklet, as candidates read it; 2013 has no accessibility edition. 5 items (7867, 11792, 15947, 33532 and 42570) ship with option_text NA because no option text could be extracted; nothing was generated to fill the gap. 7867 and 15947 have options that are drawings; 42570 has stacked-fraction options that are legible in the booklet but were not transcribed; the other 2 were not checked against the page. No English ships.
enem_2013_1mil_lc	The wording is INEP’s standard (AZUL) booklet, as candidates read it; 2013 has no accessibility edition. Thirteen items (7036, 7242, 7457, 7832, 9614, 9873, 11726, 28727, 30619, 49507, 51091, 51185 and 51201) carry a stimulus description generated for this project, each marked inline ‘(gerada por IA)’; nine are mixed, with the picture’s own printed wording transcribed outside the marked paragraph. The stimuli are comic strips, cartoons, advertisements and a museum panel, none described in the printed booklet, so these items previously shipped with only a credit and a question. Every description was checked against the booklet twice by independent reviewers. Four are medium confidence (9614, 28727, 49507, 51091), in every case because the embedded image is small rather than because what it shows is disputed. For 49507, a photographed 1911 school exercise, the two pencil annotations on the sheet are transcribed as the document’s own words and only their position is described. No English ships.
enem_2013_1mil_mt	The wording is INEP’s standard (AZUL) booklet, as candidates read it; 2013 has no accessibility edition. 4 items (31535, 41660, 43001 and 43877) ship with option_text NA because no option text could be extracted; nothing was generated to fill the gap. None was checked against the page, so whether the options are drawings or unreadable notation is not known. No English ships.
enem_2015_1mil_ch	The wording is INEP’s standard (AZUL) booklet, as candidates read it; 2015 has no accessibility edition. Seven items (8635, 24576, 28864, 37753, 45004, 53985 and 83810) carry a generated stimulus description, marked inline ‘(gerada por IA)’, four of them mixed with transcribed legend or caption text outside the marked paragraph. Every description was checked against the booklet twice by independent reviewers. 83810’s map has a four-class greyscale legend and the item asks which climate matches the weakest class, so the description names the printed class labels and assigns regions to classes; those assignments come from sampling the legend’s own grey values, not from judging by eye. 28864 is medium confidence: a twelve-panel calendar reproduced at 818x717 pixels, where some figures and animal counts are at the limit of the source. No English ships.
enem_2015_1mil_cn	The wording is INEP’s standard (AZUL) booklet, as candidates read it; 2015 has no accessibility edition. 3 items (17715, 28652 and 61065) ship with option_text NA because no option text could be extracted; nothing was generated to fill the gap. None was checked against the page, so whether the options are drawings or unreadable notation is not known. No English ships.
enem_2015_1mil_lc	The wording is INEP’s standard (AZUL) booklet, as candidates read it; 2015 has no accessibility edition. Seven items (8265, 14842, 25924, 32925, 42080, 49341 and 49858) carry a generated stimulus description marked inline ‘(gerada por IA)’, four of them mixed. Every description was checked against the booklet twice by independent reviewers. Two further items, 33156 and 83482, gained text that is not generated and carries no marker: it is wording printed inside the image itself (a poem set as a picture, a notice board), transcribed from INEP’s own page. For 49341, a Magritte painting, the description deliberately stops short of naming the relation between the figure and its reflection, because that relation is what the item asks the candidate to notice. No English ships.
enem_2015_1mil_mt	The wording is INEP’s standard (AZUL) booklet, as candidates read it; 2015 has no accessibility edition. 3 items (24019, 41659 and 49204) ship with option_text NA because no option text could be extracted; nothing was generated to fill the gap. 49204 has stacked-fraction options that are legible in the booklet but were not transcribed; the other 2 were not checked against the page. No English ships.
enem_2016_1mil_ch	The wording is INEP’s standard (AZUL) booklet, as candidates read it; 2016 has no accessibility edition. Item 54683 ships with option_text NA because no option text could be extracted; nothing was generated to fill the gap. It was not checked against the page, so whether the options are drawings or unreadable notation is not known. Four items (11802, 38717, 62021 and 88329) carry a generated stimulus description marked inline ‘(gerada por IA)’, two of them mixed with transcribed caption text outside the marked paragraph; every description was checked against the booklet twice by independent reviewers. 38717 reproduces the lettering of a Persepolis page, transcribed as the booklet prints it: the booklet prints the name ‘MARIN’, although the Brazilian edition of the book is usually cited with ‘Narine’, and the transcription follows INEP’s page. 62021 is medium confidence at 170x242 pixels. No English ships.
enem_2016_1mil_cn	The wording is INEP’s standard (AZUL) booklet, as candidates read it; 2016 has no accessibility edition. 3 items (44432, 97497 and 97711) ship with option_text NA because no option text could be extracted; nothing was generated to fill the gap. 97711 has options that are drawings; the other 2 were not checked against the page. No English ships.
enem_2016_1mil_lc	The wording is INEP’s standard (AZUL) booklet, as candidates read it; 2016 has no accessibility edition. Six items (27055, 27677, 45089, 55970, 61388 and 82375) carry a generated stimulus description marked inline ‘(gerada por IA)’, three of them mixed with the image’s printed wording transcribed outside the marked paragraph. Every description was checked against the booklet twice by independent reviewers. 45089 and 61388 are medium confidence: both are photographs of installations reproduced small enough that some background detail is indistinct, and the descriptions say only what is distinguishable at the source’s resolution. No English ships.
enem_2016_1mil_mt	The wording is INEP’s standard (AZUL) booklet, as candidates read it; 2016 has no accessibility edition. 3 items (32221, 39198 and 42955) ship with option_text NA because no option text could be extracted; nothing was generated to fill the gap. 39198’s options are diagrams, and what had been extracted was only label residue from inside the artwork; 42955 has stacked-fraction options that are legible in the booklet but were not transcribed; 32221 was not checked against the page. No English ships.
enem_2017_1mil_ch	The wording is INEP’s accessibility (Ledor, read-aloud) edition of the exam, not the printed booklet candidates read: INEP’s verbal descriptions of figures, graphs and tables are part of the text, and notation can be written out in words (‘elevado a’ for an exponent). No English ships.
enem_2017_1mil_cn	The wording is INEP’s accessibility (Ledor, read-aloud) edition of the exam, not the printed booklet candidates read: INEP’s verbal descriptions of figures, graphs and tables are part of the text, and notation can be written out in words (‘elevado a’ for an exponent). 3 items (59829, 61269 and 82421) ship with option_text NA because no option text could be extracted; nothing was generated to fill the gap. 82421 has stacked-fraction options that are legible in the booklet but were not transcribed; the other 2 were not checked against the page. No English ships.
enem_2017_1mil_lc	The wording is INEP’s accessibility (Ledor, read-aloud) edition of the exam, not the printed booklet candidates read: INEP’s verbal descriptions of figures, graphs and tables are part of the text, and notation can be written out in words (‘elevado a’ for an exponent). No English ships.
enem_2017_1mil_mt	The wording is INEP’s accessibility (Ledor, read-aloud) edition of the exam, not the printed booklet candidates read: INEP’s verbal descriptions of figures, graphs and tables are part of the text, and notation can be written out in words (‘elevado a’ for an exponent). No English ships.
enem_2018_1mil_ch	The wording is INEP’s accessibility (Ledor, read-aloud) edition of the exam, not the printed booklet candidates read: INEP’s verbal descriptions of figures, graphs and tables are part of the text, and notation can be written out in words (‘elevado a’ for an exponent). Item 96813 ships with option_text NA because no option text could be extracted; nothing was generated to fill the gap. It was not checked against the page, so whether the options are drawings or unreadable notation is not known. No English ships.
enem_2018_1mil_cn	The wording is INEP’s accessibility (Ledor, read-aloud) edition of the exam, not the printed booklet candidates read: INEP’s verbal descriptions of figures, graphs and tables are part of the text, and notation can be written out in words (‘elevado a’ for an exponent). 2 items (111411 and 111630) ship with option_text NA because no option text could be extracted; nothing was generated to fill the gap. 111411 has options that are drawings; the other were not checked against the page. No English ships.
enem_2018_1mil_lc	The wording is INEP’s accessibility (Ledor, read-aloud) edition of the exam, not the printed booklet candidates read: INEP’s verbal descriptions of figures, graphs and tables are part of the text, and notation can be written out in words (‘elevado a’ for an exponent). No English ships.
enem_2018_1mil_mt	The wording is INEP’s accessibility (Ledor, read-aloud) edition of the exam, not the printed booklet candidates read: INEP’s verbal descriptions of figures, graphs and tables are part of the text, and notation can be written out in words (‘elevado a’ for an exponent). Item 111609 ships with option_text NA because no option text could be extracted; nothing was generated to fill the gap. It was not checked against the page, so whether the options are drawings or unreadable notation is not known. No English ships.
enem_2019_1mil_ch	The wording is INEP’s accessibility (Ledor, read-aloud) edition of the exam, not the printed booklet candidates read: INEP’s verbal descriptions of figures, graphs and tables are part of the text, and notation can be written out in words (‘elevado a’ for an exponent). No English ships.
enem_2019_1mil_cn	The wording is INEP’s accessibility (Ledor, read-aloud) edition of the exam, not the printed booklet candidates read: INEP’s verbal descriptions of figures, graphs and tables are part of the text, and notation can be written out in words (‘elevado a’ for an exponent). No English ships.
enem_2019_1mil_lc	The wording is INEP’s accessibility (Ledor, read-aloud) edition of the exam, not the printed booklet candidates read: INEP’s verbal descriptions of figures, graphs and tables are part of the text, and notation can be written out in words (‘elevado a’ for an exponent). No English ships.
enem_2019_1mil_mt	The wording is INEP’s accessibility (Ledor, read-aloud) edition of the exam, not the printed booklet candidates read: INEP’s verbal descriptions of figures, graphs and tables are part of the text, and notation can be written out in words (‘elevado a’ for an exponent). Item 117635 ships with option_text NA because no option text could be extracted; nothing was generated to fill the gap. It was not checked against the page, so whether the options are drawings or unreadable notation is not known. No English ships.
enem_2020_1mil_ch	The wording is INEP’s accessibility (Ledor, read-aloud) edition of the exam, not the printed booklet candidates read: INEP’s verbal descriptions of figures, graphs and tables are part of the text, and notation can be written out in words (‘elevado a’ for an exponent). No English ships.
enem_2020_1mil_cn	The wording is INEP’s accessibility (Ledor, read-aloud) edition of the exam, not the printed booklet candidates read: INEP’s verbal descriptions of figures, graphs and tables are part of the text, and notation can be written out in words (‘elevado a’ for an exponent). 3 items (62745, 83584 and 90016) ship with option_text NA because no option text could be extracted; nothing was generated to fill the gap. 62745’s options are circuit diagrams, and what had been extracted was only label residue from inside them; the other 2 were not checked against the page. No English ships.
enem_2020_1mil_lc	The wording is INEP’s accessibility (Ledor, read-aloud) edition of the exam, not the printed booklet candidates read: INEP’s verbal descriptions of figures, graphs and tables are part of the text, and notation can be written out in words (‘elevado a’ for an exponent). No English ships.
enem_2020_1mil_mt	The wording is INEP’s accessibility (Ledor, read-aloud) edition of the exam, not the printed booklet candidates read: INEP’s verbal descriptions of figures, graphs and tables are part of the text, and notation can be written out in words (‘elevado a’ for an exponent). 5 items (30344, 49630, 60666, 111491 and 111530) ship with option_text NA because no option text could be extracted; nothing was generated to fill the gap. 60666 has stacked-fraction options that are legible in the booklet but were not transcribed; the other 4 were not checked against the page. No English ships.
enem_2021_1mil_ch	The wording is INEP’s accessibility (Ledor, read-aloud) edition of the exam, not the printed booklet candidates read: INEP’s verbal descriptions of figures, graphs and tables are part of the text, and notation can be written out in words (‘elevado a’ for an exponent). No English ships.
enem_2021_1mil_cn	The wording is INEP’s accessibility (Ledor, read-aloud) edition of the exam, not the printed booklet candidates read: INEP’s verbal descriptions of figures, graphs and tables are part of the text, and notation can be written out in words (‘elevado a’ for an exponent). 3 items (14944, 46813 and 62293) ship with option_text NA because no option text could be extracted; nothing was generated to fill the gap. 62293 has options that are drawings; the other 2 were not checked against the page. Notation that could not be read from the booklet’s font is replaced inline by a description marked ‘(gerada por IA)’ in items 54959, 59907, 85781, 111544, 117627, 117867 and 117891. No English ships.
enem_2021_1mil_lc	The wording is INEP’s accessibility (Ledor, read-aloud) edition of the exam, not the printed booklet candidates read: INEP’s verbal descriptions of figures, graphs and tables are part of the text, and notation can be written out in words (‘elevado a’ for an exponent). Notation that could not be read from the booklet’s font is replaced inline by a description marked ‘(gerada por IA)’ in items 111852 and 117572. Item 118258 also carries a generated description of its comic strip, marked inline the same way; the strip’s dialogue was already in the cell from the booklet’s text layer and is left as INEP wrote it. No English ships.
enem_2021_1mil_mt	The wording is INEP’s accessibility (Ledor, read-aloud) edition of the exam, not the printed booklet candidates read: INEP’s verbal descriptions of figures, graphs and tables are part of the text, and notation can be written out in words (‘elevado a’ for an exponent). Item 126019 ships with option_text NA because no option text could be extracted; nothing was generated to fill the gap. It was not checked against the page, so whether the options are drawings or unreadable notation is not known. Notation that could not be read from the booklet’s font is replaced inline by a description marked ‘(gerada por IA)’ in item 15209. No English ships.
enem_2022_1mil_ch	The wording is INEP’s accessibility (Ledor, read-aloud) edition of the exam, not the printed booklet candidates read: INEP’s verbal descriptions of figures, graphs and tables are part of the text, and notation can be written out in words (‘elevado a’ for an exponent). No English ships.
enem_2022_1mil_cn	The wording is INEP’s accessibility (Ledor, read-aloud) edition of the exam, not the printed booklet candidates read: INEP’s verbal descriptions of figures, graphs and tables are part of the text, and notation can be written out in words (‘elevado a’ for an exponent). No English ships.
enem_2022_1mil_lc	The wording is INEP’s accessibility (Ledor, read-aloud) edition of the exam, not the printed booklet candidates read: INEP’s verbal descriptions of figures, graphs and tables are part of the text, and notation can be written out in words (‘elevado a’ for an exponent). No English ships.
enem_2022_1mil_mt	The wording is INEP’s accessibility (Ledor, read-aloud) edition of the exam, not the printed booklet candidates read: INEP’s verbal descriptions of figures, graphs and tables are part of the text, and notation can be written out in words (‘elevado a’ for an exponent). 4 items (5961, 8364, 10409 and 63646) ship with option_text NA because no option text could be extracted; nothing was generated to fill the gap. None was checked against the page, so whether the options are drawings or unreadable notation is not known. No English ships.
enem_2023_1mil_ch	The wording is INEP’s accessibility (screen-reader) edition of the exam, not the printed booklet candidates read: notation, units and symbols are written out in words, INEP’s verbal descriptions of figures and tables are part of the text, and source credits can sit before a passage. Compared word by word with the standard booklet, only 24 of the 45 items came out identical. No English ships.
enem_2023_1mil_cn	The wording is INEP’s accessibility (screen-reader) edition of the exam, not the printed booklet candidates read: notation, units and symbols are written out in words and source credits can sit before a passage; only 5 of the 45 items came out identical to the standard booklet. Two items (78578, 54804) were transcribed from the standard booklet instead, and their figure, table and equation descriptions – plus a replacement for item 60332’s figure, which INEP declined to describe – were generated by IRW and are marked inline ‘(gerada por IA)’. Item 78578’s options are graphs, so its option_text is blank. No English ships.
enem_2023_1mil_lc	The wording is INEP’s accessibility (screen-reader) edition of the exam, not the printed booklet candidates read: notation is written out in words, INEP’s figure descriptions are part of the text, and source credits can sit before a passage; 23 of the 50 items came out identical to the standard booklet. The 50 codes cover 45 positions because the five foreign-language questions exist in an English and a Spanish version. Four items (140595, 140736, 140959, 15035) carry INEP’s ‘[supressao de texto]’ where copyrighted stimulus was withheld. No English ships.
enem_2023_1mil_mt	The wording is INEP’s accessibility (screen-reader) edition of the exam, not the printed booklet candidates read: notation, units and symbols are written out in words, and only 2 of the 44 items came out identical to the standard booklet. Two items (125902, 81742) were transcribed from the standard booklet instead, and their figure descriptions were generated by IRW and are marked inline ‘(gerada por IA)’. Item 125902’s options are diagrams, so its option_text is blank. No English ships.
enem_2024_1mil_ch	The wording is INEP’s accessibility (Ledor, read-aloud) edition of the exam, not the printed booklet candidates read: INEP’s verbal descriptions of figures, graphs and tables are part of the text, and notation can be written out in words (‘elevado a’ for an exponent). No English ships.
enem_2024_1mil_cn	The wording is INEP’s accessibility (Ledor, read-aloud) edition of the exam, not the printed booklet candidates read: INEP’s verbal descriptions of figures, graphs and tables are part of the text, and notation can be written out in words (‘elevado a’ for an exponent). Item 117859 ships with option_text NA because no option text could be extracted; nothing was generated to fill the gap. It was not checked against the page, so whether the options are drawings or unreadable notation is not known. No English ships.
enem_2024_1mil_lc	The wording is INEP’s accessibility (Ledor, read-aloud) edition of the exam, not the printed booklet candidates read: INEP’s verbal descriptions of figures, graphs and tables are part of the text, and notation can be written out in words (‘elevado a’ for an exponent). No English ships.
enem_2024_1mil_mt	The wording is INEP’s accessibility (Ledor, read-aloud) edition of the exam, not the printed booklet candidates read: INEP’s verbal descriptions of figures, graphs and tables are part of the text, and notation can be written out in words (‘elevado a’ for an exponent). 2 items (111549 and 125996) ship with option_text NA because no option text could be extracted; nothing was generated to fill the gap. None was checked against the page, so whether the options are drawings or unreadable notation is not known. No English ships.
enem_2025_1mil_ch	The wording is INEP’s accessibility (Ledor, read-aloud) edition of the exam, not the printed booklet candidates read: INEP’s verbal descriptions of figures, graphs and tables are part of the text, and notation can be written out in words (‘elevado a’ for an exponent). No English ships.
enem_2025_1mil_cn	The wording is INEP’s accessibility (Ledor, read-aloud) edition of the exam, not the printed booklet candidates read: INEP’s verbal descriptions of figures, graphs and tables are part of the text, and notation can be written out in words (‘elevado a’ for an exponent). No English ships.
enem_2025_1mil_lc	The wording is INEP’s accessibility (Ledor, read-aloud) edition of the exam, not the printed booklet candidates read: INEP’s verbal descriptions of figures, graphs and tables are part of the text, and notation can be written out in words (‘elevado a’ for an exponent). No English ships.
enem_2025_1mil_mt	The wording is INEP’s accessibility (Ledor, read-aloud) edition of the exam, not the printed booklet candidates read: INEP’s verbal descriptions of figures, graphs and tables are part of the text, and notation can be written out in words (‘elevado a’ for an exponent). Item 125905 ships with option_text NA because no option text could be extracted; nothing was generated to fill the gap. It was not checked against the page, so whether the options are drawings or unreadable notation is not known. No English ships.
enkavi_2019_ant_flanker	Item text for this cognitive task renders each trial’s visual display – the cue and the five-symbol arrow/dash array, in the task’s own notation – because the Attention Network Task presents images rather than written item wording; and resp is trial accuracy (1 = correct), not the key the participant pressed, so option_text reads Incorrect/Correct.
enkavi_2019_gonogo	The go/no-go stimuli are wordless coloured squares, so item_text is blank by design. resp is trial accuracy, not the key pressed, so option_text reads Correct/Incorrect – a scoring label rather than an option. The two colour words in the instructions were counterbalanced per participant and appear as the placeholders [go color] and [no-go color].
enkavi_2019_navon	item_text for this table is IRW’s description of the Navon stimulus, not wording anyone read: the task presented a large letter made of small letters and the deposit ships no stimulus images, so each item’s line was rendered from the task code’s own ‘ of ’ stimulus naming; resp is trial accuracy (0/1), so option_text is an accuracy label rather than a response option, and the two colours in the instructions are placeholders because the colour-to-attended-level mapping was counterbalanced across participants.
enkavi_2019_simon	The Simon stimulus is a coloured box with no text, so item_text is blank by design and the item code (colour x side x congruency) describes the stimulus. resp is trial accuracy, so option_text reads Correct/Incorrect rather than a response option. The two arrow keys were counterbalanced across participants and appear in the instructions as the placeholders [KEY FOR RED]/[KEY FOR BLUE].
enkavi_2019_stopsignal	The four stimuli are wordless shape images, so item_text is blank by design and the item code carries the shape. resp is trial accuracy, so option_text reads Correct/Incorrect rather than a response option, and there is no correct_response because the shape-to-key mapping was shuffled per session. ‘[prompt_text]’ in the instructions stands for that per-session legend, which was shown as images.
enkavi_2019_stroop	item_text for this cognitive task describes the trial’s visual display (the colour word and the ink colour it was printed in) rather than a written question, and resp is trial accuracy (1 = correct), not the key the participant pressed, so option_text reads Incorrect/Correct.
envirisk_lalot_2025	Twelve items on environmental risk. Six ask how much you agree that biodiversity loss or climate change poses a threat, on a 1-7 scale from strongly disagree to strongly agree. Four ask how acceptable you find a particular environmental policy, on a 1-7 scale from very low to very high acceptability. The last two are not attitude questions at all but political self-placements on 0-10 scales – left to right, and liberal to conservative. Each group keeps its own anchors.
environment_ltm	item_text for this table is the short item descriptor published in the ltm R package’s Environment documentation (e.g. ‘Lead from petrol.’), not the verbatim question stem read to respondents in the 1990 British Social Attitudes Survey, which the package does not reproduce; no instruction text is available for the same reason.
esiason_2024_ace	Item wording is the standard 10-question Adverse Childhood Experiences handout; the study itself published no wording and cites the clinical ACE-Q (Murphy et al., 2016), whose exact phrasing sits in a paywalled book chapter and could not be compared. The handout’s multi-part questions are flattened onto a single line each, so an item here can carry what the printed form lays out as several clauses.
esiason_2024_cfq	Item wording is the canonical CFQ13 (Gillanders, Bolderston & Bond); the study itself published no item text, and the response anchors shipped are the instrument’s own (‘never true’ … ‘always true’) rather than the paper’s paraphrase ‘1 (not true at all) to 7 (always true)’.
estcrm_selfeff	The item wording is the one-line English description of each column in the EstCRM R package documentation, not a transcript of the administered form. The scale (Kan 2009) comes from a Turkish study and was probably administered in Turkish, but no source states the language, so the English is likely a rendering rather than what respondents read.
estevezlopez_2016_panas	The PANAS was administered in Spanish (Sandin et al., 1999 adaptation), but neither the paper nor its data deposit publishes the Spanish wording, so the item text shipped here is the English affect terms and response anchors printed in the paper, not the sentences respondents read.
estrella_2023_psq18	The study adapted the PSQ-18 instructions to ask about medical care for joint hypermobility, where RAND’s form asks about medical care in general. resp is stored as printed, 1 = strongly agree to 5 = strongly disagree, for every item: the nine items RAND reverse-scores (psq18_1, 2, 3, 5, 6, 8, 11, 15, 18) are not reversed, so a higher value means less satisfaction on those items and more satisfaction on the other nine.
evpromisi_stone_2021_cdiag	The IRW dictionary describes this table as PROMIS fatigue, but its 12 items are the study’s own baseline chronic-diagnosis checklist (heart disease, diabetes, cancer, …) asked once at enrolment, not a PROMIS item bank; the item text shipped here is that checklist, transcribed from the study codebooks.
extremera_2016_ei	The Spanish item wording is the authors’ own WLEIS-S adaptation, but the study’s SPSS deposit carries no variable labels and the paper reproduces no items, so the assignment of the 16 sentences to codes ei1..ei16 is an inference from the administration form’s printed order; it is confirmed at the subscale level against the data (within-subscale r .53 vs .40 between, which rules out the alternative blocked numbering the same authors use in Psicothema 2019) but the order of the four items within each subscale is not independently verified. The English in option_text_translated for the five intermediate anchors was generated by IRW, not published by the authors, who state only the 1 = strongly disagree / 7 = strongly agree endpoints. The item code ei16 is IRW’s normalisation of the source file’s own ie16, a naming slip in the raw SPSS data; the other fifteen codes are the source column names unchanged.
extremera_2016_sbq	Item text is the English SBQ-R; the study administered its own unpublished professional Spanish translation, which is not reproduced in the paper or its SPSS deposit. resp values are the questionnaire’s raw printed option numbers (item 1: 1-6, item 4: 0-6), not the SBQ-R’s collapsed 3-18 scoring – summing them does not give a standard SBQ-R total.
FACIT_YOUNT_2021_clinic10	The deposit publishes no question stem for this medical-conditions checklist, so the item text shown here is the data file’s own variable labels (for example “Ever had condition - Anxiety”) rather than the sentence respondents read; the instructions and block prompt are blank for the same reason. The wording was never deposited – the SPSS labels, the codebook PDF and the data dictionary were all checked. Item codes and response coding are exact: source-to-live per-item by per-level counts match in all 36 cells.
fad_dataset1	The item text here is the administered Chinese, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. Published English exists for the FAD-Plus, but the Chinese item set here does not map one- to-one onto it, so the administered Chinese is translated instead.
fad_dataset2	The item text here is the administered Chinese, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. As fad_dataset1: the published FAD-Plus English does not map one-to-one onto this item set. The ten items this table has beyond fad_dataset1 are translated on the same basis.
FAD_fadplus_goto2021	The 54 item codes are 27 items, not 54: the same 27 FAD-Plus items appear in two different Japanese translations, both given to the same respondents – FAD_G_* is the FAD-J (Goto et al. 2015) and FAD_W_* is the FAD+ (Watanabe et al. 2014), although the table name mentions only Goto. item_text is the administered Japanese; the _translated column holds the English FAD-Plus original from the study’s supplementary table, not a translation of the Japanese.
falih_2026_dass21	The Falih deposit publishes no item wording at all, so the text shown is the canonical English DASS-21 taken from the instrument’s own distribution site; the study was run at the University of Baghdad and may have administered an Arabic translation, which the deposit does not document. Which DASS-21 item each code carries is pinned to the subscale block by the deposit’s own DEPRESSI/ANXIETY/STRESS column layout (an exact match to the canonical partition), but the order of the seven items within each block rests on the code numbering rather than on any published item-level statistic.
falih_2026_mds16	The order of items within each of the scale’s two factor blocks rests on the source file’s column order rather than on anything in the data: the published two-factor structure separates cleanly here, but swapping two items inside the same block would leave every statistic unchanged.
famous_melodies	Five things listeners rated about each of 109 well-known melodies: how positive or negative it sounds, how calming or exciting, how familiar it is, roughly what age they learned it, and whether they could name it. NOTE THAT THE ROWS ARE RATINGS OF A MELODY – the melody is the id and the listener is the rater, the reverse of a questionnaire. The scales differ by dimension (naming is yes/no, age runs 1-8, the rest 0-4) and the labels for the scale points are not published, so the direction of each rating cannot be read off this table.
fang_2021_pbc	The item text is the English wording printed in the article’s questionnaire tables, the authors’ rendering of a questionnaire given in Chinese to grade 5-6 pupils in Taiwan; the Chinese wording respondents read is not published.
fang_2021_peb	As for fang_2021_pbc: the article’s English wording, not the Chinese wording respondents read.
fang_2021_pn	As for fang_2021_pbc: the article’s English wording, not the Chinese wording respondents read.
fang_2026_anime_museum	The English item text (item_text_translated) was written by IRW from the Chinese questionnaire stems respondents read; it is not the authors’ wording.
fauziah_2026_human_ai_collab	As for fauziah_2026_humble_leadership: IRW’s English translation of the Indonesian headers.
fauziah_2026_humble_leadership	The English item text (item_text_translated) was written by IRW from the Indonesian form headers respondents read; it is not the authors’ wording.
fauziah_2026_intrinsic_motivation	As for fauziah_2026_humble_leadership: IRW’s English translation of the Indonesian headers.
fauziah_2026_iwb	As for fauziah_2026_humble_leadership: IRW’s English translation of the Indonesian headers.
fauziah_2026_psych_safety	As for fauziah_2026_humble_leadership: IRW’s English translation of the Indonesian headers.
FEDSP_Trzcinska_2023_MonKnow	item_text is IRW’s bracketed description of what the preschoolers were shown (‘[children were shown: …]’), not wording anyone read or heard: the spoken question put to the child is not published in the deposit and the paper is paywalled. The zloty amounts come from the study’s data file; the framing and the coin/note distinction are IRW’s. The task was administered in Polish, but the description is in English.
fernandez_prados_2025_ai_uses	As for fernandez_prados_2025_attari: IRW-written English from the Spanish questionnaire, not the instrument’s published English wording.
fernandez_prados_2025_aias	As for fernandez_prados_2025_attari: IRW-written English from the Spanish questionnaire, not the instrument’s published English wording.
fernandez_prados_2025_attari	The English translations (*_translated) were written by IRW from the Spanish questionnaire respondents read. They are not the study’s own, and they are not the instrument’s published English wording.
fernandez_prados_2025_digcomp_comm	As for fernandez_prados_2025_attari: IRW-written English from the Spanish questionnaire, not the instrument’s published English wording.
fernandez_prados_2025_digcomp_info	As for fernandez_prados_2025_attari: IRW-written English from the Spanish questionnaire, not the instrument’s published English wording.
fernandez_prados_2025_gaais	As for fernandez_prados_2025_attari: IRW-written English from the Spanish questionnaire, not the instrument’s published English wording.
Fh_Okcsr_Roos_2022_study1_Feeling_Heard	Items Q6.2_1-Q6.2_4 are COVID-19 emotion-frequency questions rather than feeling-heard items, and items ending _16 are attention checks. For the Q4/Q5 attention checks the English reads ‘select Agree’, following the Dutch labels and the responses, where the study codebook’s English says ‘Disagree’.
fisher_2019_belonging	The 1-6 response anchors are not shipped: the study’s only supporting file is a numeric data spreadsheet with no value labels, and the survey instrument the paper refers to was not deposited, so the option wording and scale direction are undocumented by the source.
fisher_2019_structure	The 1-7 response anchors are not shipped and no claim is made about scale direction: the paper publishes both item stems verbatim but never states the anchors, and the survey materials its Measures section promises were not deposited – the only supplement is the numeric data spreadsheet. This is a gap the source never filled rather than a mismatch between the text and the table.
fivpei_perrig_2023_attdiff	Paper has 28 items, IRW contains only 21
florida_twins_friends	Items friends20/friends21 are not documented in the Florida Twin Project’s Wave 1 Child Survey Measures codebook; the other 19 items match exactly
florida_twins_par	Two of the 31 codes are not questions: parstrict and parwarmth are the study’s own sum scores over the other items, so they carry a bracketed IRW note in place of a stem and no option text. par6 is stored raw with ‘As late as I want’ folded to 0, which is not the recode the codebook prints for it.
FomoNegativeAffect_cremer_2026_panas	Despite the table name, this table holds all 20 PANAS items (positive and negative affect), not only the negative-affect items the paper analysed. The item wording and middle anchors are the standard PANAS form; the paper confirms only the endpoints, and the study’s past-month instruction stem is unpublished, so no instructions are shipped. The paper’s ‘fear’ (six items) and ‘distress’ (four items) subscales correspond to items 2, 4, 6, 8, 11, 13 (distressed, upset, guilty, hostile, irritable, ashamed) and 7, 15, 18, 20 (scared, nervous, jittery, afraid), which by content look reversed.
FomoNegativeAffect_cremer_2026_phq	The study administered the PHQ-9 with a past-month reference period rather than the standard “Over the last 2 weeks”, and did not publish its stem, so no instructions are shipped; item and option wording is the standard English PHQ-9. The deposit carries no item labels: which items form the somatic (3, 4, 5) and cognitive blocks is confirmed by reproducing the deposit’s own subscale sums (461/461), but the order within each block rests on the standard PHQ-9 numbering.
fox_2023_bct	Item BCT_14_Climate is shipped with Study 1’s wording (‘it is in their interests to do so’); Study 2 respondents (cov_study = 2) read ‘it is in their financial interests to do so’.
FPAS_Silva_2022	The Portuguese item wording is the study’s REDCap data-dictionary labels, which differ slightly on most items from the FPAS-Br printed in the paper’s Table 1; which version respondents read is not established. The English is the original FPAS wording (Burns et al., 2005) printed alongside in the paper, not a translation of the Brazilian adaptation, and it departs from it on items 3, 7 and 14. The response coding is reversed on half the items (items 1-4, 12, 14, 15, 17 and 18 store 5 = Discordo totalmente), as the option text shows.
fraijo_2022_mslq	Item wording is the study authors’ English rendering of a questionnaire administered in Spanish (the Spanish was not deposited), transcribed with the source’s typos (items 3, 4, 11, 30, 51, 57). The eight negatively worded items (item33, 37, 40, 52, 57, 60, 77, 80) are stored raw, not reverse-scored, and the deposit’s own subscale scores were computed the same way.
franco_2024_unfolding	Six statements about how much value the respondent places on receiving attention from others, ranked from the one that describes them best (1) to the one that describes them worst (6) – so the response is a rank, not a rating, and every respondent uses each number once. The statements were generated by ChatGPT for a tutorial on unfolding models and administered in Brazilian Portuguese, shown here with the authors’ English alongside. They are ordered from the most to the least attention-seeking.
fredrickson_2015_mhcsf	The item wording, anchors and instructions are the canonical Keyes MHC-SF form, not the study’s own materials, which reproduce none. The shipped instructions are the paper form’s, including ‘Place a check mark in the box’, while this study administered the questionnaire online. Item 6 is given in the current wording (‘that our society is a good place, or is becoming a better place, for all people’); Keyes also documents an original wording, and the study does not say which its participants saw.
ftna_kasper_2022	The “items” in this table are whole-subject examination grades (Mathematics, Kiswahili and so on), not individual exam questions. Note also that the three core subjects reuse one item code across both waves, even though wave 0 is the primary-school PSLE and wave 1 the secondary-school FTNA – so the same code refers to a different examination in each wave
fukuda_2021_health_literacy	Two things to know before scoring this table. The response scale runs the opposite way from the published HLS-EU-Q47 key and from the paper’s own Methods section: the deposited data codes 1 = very easy through 4 = very difficult, so a HIGHER value means MORE difficulty and LOWER health literacy, and the shipped option_text follows the data rather than the paper. And the table holds 46 of the instrument’s 47 items – hl_item39 through hl_item46 are HLS-EU-Q47 questions 40 to 47, because question 39 was dropped when the source file was processed, so codes above the gap do not equal the canonical item numbers. Item text is the canonical English from the HLS-EU Consortium’s own annex; the survey was administered in Japanese, which neither the paper nor its deposit publishes.
fukuda_2021_info_reliability	The survey was administered in Japanese to educators in Japan, but the study’s deposited data file and paper publish only English labels – for the eight information sources and for the 1-5 reliability anchors alike – so the wording shown here is the authors’ English and the Japanese respondents actually read is not recoverable from the source.
fukuda_2021_withholding_behavior	The six item texts are the English questions printed in the source paper’s Table 4, fuller than the terse column headers the data file carries; the survey itself was administered in Japanese and neither the paper nor its deposit publishes that wording, so the translated columns are empty by design rather than pending.
gabriel_2026_knowledge_confidence	Item text here is the authors’ published English rendering of statements administered in German; the German originals appear in neither the deposit nor the paper, so only the response anchors ship in the language respondents saw, with the authors’ English alongside them.
gabriel_2026_knowledge_correct	Despite its name, this table stores each respondent’s true/false judgement (1 = judged true, 0 = judged false), not whether the answer was correct; score it against correct_response to get correctness, which reproduces all nine published correct rates. Item wording is the authors’ English rendering of statements administered in German; the response options are the administered German (Richtig/Falsch).
gabriel_2026_media_use	Item wording is the authors’ own English rendering rather than the German the survey was administered in, and the shared question stem is recorded once in instructions since this was a matrix question; the response anchors, unlike the items, are the administered German, with the authors’ English in option_text_translated.
gad_BrummerHoffman_2021	This table pools two administration arms whose wording differed by design: the text shown is the Study 1 grid form, with the shared frame in instructions and the symptom alone in item_text, while the 572 Study 2 respondents saw each item as one complete Portuguese sentence. The English translated columns are mixed provenance – symptom phrases come from the deposit’s own English questionnaire, but the stems and most anchors were translated for IRW, because the deposit’s English form prints canonical GAD-7 anchors that do not correspond to the Portuguese actually administered.
gan_2015_cesd	The study administered a Chinese CES-D, but neither the article nor its deposit publishes any Chinese wording, so the text shown is Radloff’s (1977) original English. Note also that responses are stored raw rather than reverse-scored: the four positively worded items (CES4, CES8, CES12, CES16) are not reversed in the table, so the shipped frequency anchors read in the same direction for all twenty items and anyone scoring a total must reverse those four themselves.
gan_2015_ucla_loneliness	The item text is the English ULS-8 wording; the study administered a Chinese translation that neither the paper nor its deposit reproduces. The deposited responses are already reverse-scored, so for UCLA3 and UCLA6 – the two items the instrument’s own scoring key reverses – the shipped anchors run 1 = Always to 4 = Never, opposite to the other six.
gan_2024_depression	Item text is the canonical English HDRS-17 form; this study administered the 1988 Chinese version and no Chinese wording is published in the deposit or paper, so the English is a substitute rather than the wording raters used. 抑16 and 抑162 are the HDRS’s two alternate ratings of item 16 (loss of weight: by patient report vs by weekly measurement); both were scored for every patient and both count in the study’s weight factor and total.
ganbat_2022_pollution_disease_risk	The five items here are the study’s derived disease categories rather than anything respondents read: the questionnaire’s Question 2 offered sixteen named diseases as a check-all list, and the released data aggregates them into Respiratory tract, Cardiovascular, Cerebrovascular, Maternal and baby outcomes related, and Diabetes.
ganbat_2022_pollution_leave_type	Q7 is a check-all-that-apply battery: resp 1 means the respondent selected that leave type and 0 means they did not, and since the questionnaire prints no labels for that coding, option_text is deliberately blank rather than missing. Item Q7.5 is an “others” residual that appears only in the dataset codebook – the questionnaire itself lists four leave types – so no administered Mongolian wording exists to ship for it, and its item_text is blank by the same principle.
ganbat_2022_pollution_symptoms	Each item is a binary flag covering a group of adjacent checkbox options rather than one question, so item_text lists the constituent administered options – Q1.5, for instance, covers blurry, sore, red, itchy and runny eyes. The questionnaire’s “Chest pain” option has no column in the released data at all.
gao_2022_emotional_resilience	The item and instruction text is the administered Chinese from the study’s data file. The English in the _translated columns was produced by IRW; no English wording exists in the deposit or the paper. Response options are unlabelled because the paper describes a seven-point 0 (never) to 6 (always) scale while the data hold six levels (1-6). Whether the negatively worded items 2, 5, 7, 10 and 11 are stored raw or reversed cannot be determined from the source.
gao2025_attachment_anxiety	The item wording for this table is the English questionnaire in the study’s own deposit; the study was administered in Chinese and no Chinese wording was deposited. The assignment of the three anxiety statements to CAnxiety1-3 follows the questionnaire’s item order and could not be verified item by item from the data.
gao2025_spiritual_wellbeing	Item text for this table is the English questionnaire wording deposited by the study; the instrument was administered in Chinese to older adults in Henan Province and no Chinese wording appears in the deposit, so these are not the strings respondents read
garciabatista_2021_erq	The Emotion Regulation Questionnaire was administered in Spanish to Dominican health workers, but no Spanish wording is published in the study’s deposit or supplements, so the item text shipped here is the authors’ own English rendering taken from their SPSS variable labels; response-option labels are blank because the administered 5-point scale’s anchors are not published anywhere in the source (the paper’s Methods describes the original 7-point format, which is not what was administered).
gcbs_brotherton_2013	The item text is the wording of the online form at openpsychometrics.org, fetched in 2026; the data were collected in 2016 and it is not confirmed that the page was identical then. That form differs from the published scale (Brotherton et al., 2013, Table A1): Q7 reads ‘all major decisions’ where the published scale has ‘all major world decisions’, Q13 reads ‘rumours’, and the 1-5 scale was labelled only ‘Disagree’, ‘Neutral’ and ‘Agree’ at points 1, 3 and 5 rather than ‘Definitely not true’ to ‘Definitely true’. resp 0 is not a response option: it is the source file’s code for an item left unanswered (108 cells) and carries no label.
gcbs_brotherton_2013_tipi	This table is the Ten Item Personality Inventory administered after the Generic Conspiracist Beliefs Scale in the same online session. resp 0 (149 responses, 13-17 per item) is the source file’s code for an unanswered item, not a point on the 1-7 scale, and carries no label. Responses are stored raw: the reverse-keyed TIPI items (2, 4, 6, 8 and 10) are not reversed.
gcbs_brotherton_2013_vcl	A sixteen-word vocabulary check-list: respondents ticked every word whose definition they were sure they knew. Three of the words – cuivocal, florted and verdid – are not real words, and were included so that over-claiming could be detected; they are marked as such here. resp is 1 for checked and 0 for unchecked. The list was given alongside a conspiracy-belief scale as a validity check.
genpsych_russell_2024_gemma	These 28 items are not a canonical personality inventory: they were generated de novo by Gemma-2 under the AI-GENIE procedure (Russell-Lasalandra, Christensen & Golino 2024) and administered to ~1001 Prolific respondents. Item text, instructions and option labels are verbatim from the study’s own Qualtrics export, but the instrument name shown here is a descriptive label written by IRW, not a published title. The four sibling tables (gpt3_5, gpt4o, llama3, mixtral) use the SAME items_N codes for a DIFFERENT, model-specific item pool – item text must never be reused or joined across them.
genpsych_russell_2024_gpt3_5	These 31 items are not a canonical personality inventory: they were generated de novo by GPT-3.5 under the AI-GENIE procedure (Russell-Lasalandra, Christensen & Golino 2024) and administered to ~1001 Prolific respondents. Item text, instructions and option labels are verbatim from the study’s own Qualtrics export, but the instrument name shown here is a descriptive label written by IRW, not a published title. The four sibling tables (gemma, gpt4o, llama3, mixtral) use the SAME items_N codes for a DIFFERENT, model-specific item pool – item text must never be reused or joined across them.
genpsych_russell_2024_gpt4o	These 35 items are not a canonical personality inventory: they were generated de novo by GPT-4o under the AI-GENIE procedure (Russell-Lasalandra, Christensen & Golino 2024) and administered to ~1001 Prolific respondents. Item text, instructions and option labels are verbatim from the study’s own Qualtrics export, but the instrument name shown here is a descriptive label written by IRW, not a published title. The four sibling tables (gemma, gpt3_5, llama3, mixtral) use the SAME items_N codes for a DIFFERENT, model-specific item pool – item text must never be reused or joined across them.
genpsych_russell_2024_llama3	These 30 items are not a canonical personality inventory: they were generated de novo by Llama-3 under the AI-GENIE procedure (Russell-Lasalandra, Christensen & Golino 2024) and administered to ~1001 Prolific respondents. The wording carries the model’s own grammatical slips (“others opinions”, “friends backs”) and inconsistent terminal punctuation; these are transcribed verbatim from the study’s Qualtrics export and are not transcription errors. The instrument name shown here is a descriptive label written by IRW, not a published title. The four sibling tables (gemma, gpt3_5, gpt4o, mixtral) use the SAME items_N codes for a DIFFERENT, model-specific item pool – item text must never be reused or joined across them.
genpsych_russell_2024_mixtral	These 32 items are not a canonical personality inventory: they were generated de novo by Mixtral under the AI-GENIE procedure (Russell-Lasalandra, Christensen & Golino 2024) and administered to ~1001 Prolific respondents. Item text, instructions and option labels are verbatim from the study’s own Qualtrics export, but the instrument name shown here is a descriptive label written by IRW, not a published title. The response direction is pinned by the data rather than assumed: exactly 2 of the 32 items (items_6, items_14) never take the lowest response, fixing Strongly Disagree=1. The four sibling tables (gemma, gpt3_5, gpt4o, llama3) use the SAME items_N codes for a DIFFERENT, model-specific item pool – item text must never be reused or joined across them.
geography	Each item is a place on an outline map rather than a questionnaire item, and its item_text is the place name from the source archive’s own lookup table with the place type appended in parentheses – an annotation added here, not part of the source name – because 46 place names are shared by two different places. The task presented no fixed answer options, so option_text labels the 0/1 correctness score rather than transcribing anything the learner saw
gerber_2022_altruism	The item text shown here is the English wording of the Witt & Boleman adaptation of Rushton’s Self-Report Altruism Scale; the study administered its own French translation, which it does not publish. One item is not a quotation: the source that documents the adaptation prints only 13 of its 14 adapted items, so alt2’s wording was reconstructed by IRW from the dropped-item column of that table. Which statement each of alt1-alt14 refers to follows the instrument’s published item order; the data pin alt8 (lowest corrected item-total correlation, 0.368 against 0.492 for the next lowest, matching the same item’s lowest published loading in an independent validation) and the three floor and two ceiling items, but not the mid-range items alt3, alt4, alt10 and alt11, whose means sit within 0.27 of one another.
gerber_2022_eas_temperament	The children answered a French self-report version of the EAS temperament questionnaire; neither the paper nor its deposit publishes the French wording, so the item text shown here is an English rendering of the same numbered EAS items taken from another study (IJERPH 2022;19(3):1387, Table 2) and is third-person where the children read first-person. Read SOC4 with care: its English here, ‘Is lonely’, renders an item that in the original describes a child who tends to be solitary. Six items (ACT2, ACT5, TIM2, TIM3, TIM5, SOC4) are stored reverse-recoded in the IRW table, so their response anchors run 1 = strongly agree to 5 = strongly disagree, the opposite direction from the other fourteen.
germann_2026_environment	env1 (the fracking statement) is stored reverse-coded so that higher values mean more support for environmental protection: resp 1 = Completely agree, 5 = Completely disagree.
germann_2026_immigration	imm1 and imm3 are stored reverse-scored relative to their displayed pro-immigration statements (1 = Completely agree … 5 = Completely disagree), while imm2 runs 1 = Completely disagree … 5 = Completely agree, so option text differs by item.
germann_2026_redistribution	redistr2 and redistr3 are stored reverse-scored (1 = Completely agree, 5 = Completely disagree) so that higher values mean more support for redistribution on all four items; their option text reflects that direction.
germann_2026_state_intervention	Which of statint1 and statint2 is the railways item and which the NHS item is inferred from the order in the authors’ script and the supplement’s table, and cannot be checked independently against the data; the item wording, direction (1 = Completely disagree … 5 = Completely agree, no reversal) and scale statistics are verified.
gesbert_2021_tdeq	Item wording is the administered French; the response scale is published only as English endpoint labels (1 = strongly disagree, 6 = strongly agree), so option_text is in English and the four intermediate points 1<resp<6 carry no label because the source prints none.
ghanbari_2016_helma_access	Two of the eleven items, access10 and access11, ship with blank item text: they were dropped during validation of the HELMA and their wording appears in neither the published questionnaire nor the paper. Note also that this table is the pre-final field-test ‘access’ block, not the final HELMA access subscale – access1-access4 became the published instrument’s self-efficacy subscale (items 1-4) and access5-access9 its access subscale (items 5-9).
ghanbari_2016_helma_appraise	Respondents completed the 47-item pilot pool, while the wording shipped here is the 44-item final HELMA form deposited with the paper; the five appraisal items are unchanged between the two versions, which is what the PCA loading replication confirms.
ghanbari_2016_helma_comm	The item-to-text alignment is inferred from the order items appear in the source, not from an explicit code-to-text mapping
ghanbari_2016_helma_numeracy	One sentence of this table’s Persian text - the explanation of the body-mass-index formula that precedes items 43 and 44 - has no English counterpart in the study’s own English questionnaire, so its translation was generated by IRW rather than by the authors.
ghanbari_2016_helma_reading	This table is named for the HELMA reading subscale but holds the pre-final 47-item field-test form’s six-column reading block: reading1-reading5 are the published reading subscale (questionnaire items 10-14), while reading6 is questionnaire item 15, which the paper’s own factor analysis places in the understanding subscale.
ghanbari_2016_helma_understand	This table holds nine of the ten items of the HELMA understanding subscale: respondents completed the 47-item pilot form, whose understanding block has nine columns, and the subscale’s tenth published item (‘I can easily understand the meaning of the signs used in hospitals and medical centres’) was a reading-block pilot item that loaded on understanding, so it appears in the reading table instead. The item wording shipped here is the 44-item final questionnaire’s; the codes understand1 and understand2 could not be told apart in the data, so if a swap exists anywhere it is between those two items.
ghanbari_2016_helma_use	The deposited responses come from the pre-final 47-item HELMA field-test form, so this table has five items where the published scale’s use subscale has four: use1-use4 are final questionnaire items 30-33, and use5 is an item dropped during the factor analysis whose wording the article never prints, so it ships with no item text.
gholami_2017_periodontal_knowledge	The live resp for this table is the study’s own correct/incorrect scoring (0/1) of a four-option multiple-choice question, so option_text reads False/True rather than the four answer alternatives; those are printed in the paper’s Methods, the correct one being ‘soft, colourless and sticky deposits on the teeth that contain microbial and food debris’ (Q1), ‘dental plaque’ (Q2) and ‘red gingiva’ (Q3). The wording ships in English although the questionnaire was administered in Persian.
gilbert_meta_112	Only three of the eight items (1, 4 and 5) carry their question. The study used two parallel versions of the test, which ask different questions at positions 2, 3, 6, 7 and 8, and the table does not record which version a child sat, so those codes ship with blank item_text. The test is in Dutch; the English is IRW’s translation. Item 1 is five figures sharing one option list scored as a single 0/1 item. resp is accuracy only: the option a child chose is not in the data and no answer key was published, so the option rows carry no resp.
gilbert_meta_13	This table pools two different test forms (baseline and midline) under one set of item codes, and 12 of the 37 codes carry a different question at each wave, so their item_text is blank; do not assume a code means the same question at both waves. Arithmetic items read e.g. ‘12 + 6 = 18’ because the enumerator-facing form printed the answer; the child saw the problem alone. The object-identification and number items were pictures, so item_text is blank and the expected answer is in correct_response.
gilbert_meta_14	This table pools two different test forms (baseline and midline) under one set of item codes, and 39 of the 69 codes carry a different question at each wave (e.g. addition_q2 is ‘6 + 1 = 7’ at baseline and ‘0 + 3 =’ at midline; wordprob1 and wordprob2 swap), so their item_text is blank. Arithmetic items read e.g. ‘12 + 6 = 18’ because the enumerator-facing form printed the answer; the child saw the problem alone.
gilbert_meta_16	For 37 of the 71 items the source replication package gives two different renderings of the question: the shipped item_text is the study’s final India-adapted item map (e.g. q3016 ‘About how many biscuits were sold on Friday?’), while the Stata data files’ own variable labels carry shorter, apparently pre-adaptation wording for the same code (q3016 ‘sandwiches’, q3024 ‘Chris’ for ‘Krishna’, q4009 ‘Kendra begins her homework’ for ‘Kamala begins her paper route’). The item code to item identity mapping is unaffected and identical in both; only the exact sentence a student read is uncertain for those items
gilbert_meta_27	This table pools the baseline and endline waves of the ASER mother language test, and the two waves used different letters, words and reading passages, so each item’s text gives the study’s own question label followed by BOTH waves’ stimuli, marked baseline/endline; the story and the village-name item were administered at endline only. Response 1 means the surveyor coded ‘haan’ (yes/succeeded) and 0 means ‘nahin’ – the source form’s third option, ‘koi jawab nahin’ (no response), was recoded to 0 by the study’s own genvar.do. Three items (maser_lang_lttrs, maser_lang_word1, maser_lang_word2) were counts in the original form (0-10 letters read, 0-5 words read) and reach IRW as 0/1 through a dichotomisation made upstream whose threshold is not documented anywhere in the deposit, so those three carry no response-option labels. The English for the paragraph and story passages was translated by IRW; every other English string is the study’s own.
gilbert_meta_28	Seventeen tasks from a numeracy and mobile-phone test given orally to mothers, covering number reading and identification, addition and subtraction, word problems, telling the time, and using a phone. The wording shown is the study’s own short description of each task rather than what an enumerator read out. THE TEST CASCADES – two-digit addition was only attempted by mothers who passed one-digit addition – so the harder items have far fewer respondents and their pass rates are conditional on an already more able group. KNOWN PROBLEM WITH ONE ITEM: maser_math_read1 (reading digits 1-9) passes at under 2 per cent where identifying the same digits passes at 55 per cent, which is not possible; it appears to be dichotomised one step too low, the same fault already reported for the child version of this test. Its item text is correct; treat its resp as unreliable.
gilbert_meta_29	This table pools the baseline and endline waves of the ASER child language test, and the two waves used different letters, words and reading passages, so each item’s text gives the study’s own question label followed by BOTH waves’ stimuli, marked baseline/endline; the story item was administered at endline only. Response 1 means the surveyor coded ‘haan’ (yes/succeeded) and 0 means ‘nahin’ – the form’s third option, ‘koi jawab nahin’ (no response), was recoded to 0 by the study’s own genvar.do. Three items (caser_lang_lttrs, caser_lang_word1, caser_lang_word2) were counts on the original form (0-10 letters read, 0-5 words read) and reach IRW as 0/1 through a dichotomisation made upstream whose threshold is not documented anywhere in the deposit, so those three carry no response-option labels. The English for the paragraph and story passages was translated by IRW; every other English string is the study’s own.
gilbert_meta_30	item_text is the study’s own label for each task, not a fixed question: the enumerator picked the digits and numbers to ask from a printed pool, and the baseline and endline sheets print different numerals. resp is 1 only for a perfect count or a pass, not the count the study recorded. caser_math_read1 appears to be dichotomised one step too low (1.3% pass rate where the study’s coding implies 58-64%), so treat its resp as unreliable.
gilbert_meta_35	item text covers 8 of the 12 values_* items. The other four (values_b, values_h, values_k, values_l) are omitted: the study’s replication do-file builds its democratic-values scale from only eight of the twelve, and no wording is published for the other four. The eight are mapped to their text using the source .dta’s variable labels, which name six of them outright; note that values_e and values_j are reverse-worded, so their response options run in the opposite direction to the other six
gilbert_meta_38	survey does not contain bought_ins
gilbert_meta_42	item text covers only 12 of the 20 items – the 4 life_satisfaction_* and 4 locus_of_control_* items are omitted because no item text is available for them. The paper names the source instruments (a 4-item subset of Diener’s Satisfaction With Life Scale, and 4 items from the Levenson Multidimensional Internal Locus of Control Scales, both adapted for the study context) but neither it nor its appendix gives which items were used or their adapted wording, and the openICPSR questionnaire is access-gated
gilbert_meta_45	Item text for the 22 Maslach Burnout Inventory items is withheld on licence grounds; only the MBI response anchors ship. The 8 Personal Accomplishment items are already reverse-coded in the table, the opposite of the direction respondents answered in, so ‘Never’ maps to resp 6 and ‘Every day’ to resp 0 on those items; the option rows reflect the reversal.
gilbert_meta_51	ch_anemic is not an item in the code book
gilbert_meta_55	The endline assessment was administered orally in Bangla, but the deposit and both working-paper appendices publish only the authors’ English rendering of the questions, so the item text shown here is English rather than what children actually heard. Each question also came in three grade-level variants (Grade 1, Grade 2, Grades 3-4) and the table records no grade, so all three wordings are shown for every item.
gilbert_meta_64	The item text here is the administered Arabic, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. Item texts embed the template variable ${fc_name}, kept verbatim in both columns – it is a placeholder the instrument fills with the child’s name, not text to translate. Levantine Arabic forms appear alongside standard Arabic.
gilbert_meta_68	Item text is the study’s own English task label (e.g. ‘Comprehension’, ‘Word dictation’), not the Hindi question a child read: the written Hindi test came in parallel samples and the table records no sample, so items within a family share one label (eight read ‘Comprehension’, four ‘Word dictation’). The Haryana data label items only by question number, so the labels are bridged from the study’s dictionary for its Bihar/Uttarakhand arm by matching question numbers – an inference resting on Haryana having used the same forms. Only sample 1 of the Haryana test is deposited, so its Hindi wording is deliberately not shipped. resp is 0/1 with 1 = correct inferred from the deposit’s scoring code; option_text is blank.
gilbert_meta_69	Item text is the study’s own English task label (e.g. ‘Multiplication’, ‘Addition(2-3 digit)’), not the specific sum a child was given: the written mathematics test came in parallel samples and the table records no sample, so several items share one label (five read ‘Multiplication’, four ‘Addition(2-3 digit)’). The Haryana data label items only by question number, so the labels are bridged from the study’s dictionary for its Bihar/Uttarakhand arm by matching question numbers – an inference resting on Haryana having used the same forms. Only sample 1 of the Haryana test is deposited, so its wording is deliberately not shipped. resp is 0/1 with 1 = correct inferred from the deposit’s scoring code; option_text is blank.
gilbert_meta_70	The study used four parallel forms and the table does not record which one a child took, so item_text for the codes whose content differs between forms is the task the item asks (for example ‘Encircle the word which matches with the picture’), not the letter, word or passage a child saw, and several codes share it. Twelve of the 33 codes print the same content on every form and ship it. Nine items are a picture with no printed words and the dictation items were read aloud, so those carry only the task. The test is in Hindi; the English is IRW’s translation. resp 1 = correct is inferred from the study’s scoring code, not labelled, and no answer key was published.
gilbert_meta_71	The study used four parallel forms and the table does not record which one a child took, so for 12 of the 20 codes no single problem belongs to the code and item_text is blank; the 8 positions that print the same problem on all eight booklets ship it. The test is in Hindi. resp 1 = correct is inferred from the study’s scoring code, not labelled, so option_text is blank.
gilbert_meta_73	item_text is the task each item asks (‘Multiplication’, ‘Solve the sums’), not the problem a child saw: the study used four parallel forms with different numbers and the table does not record which form a child took, so several codes share the same text. resp 1 = correct is inferred from the study’s scoring code, not labelled, so option_text is blank.
gilbert_meta_95	Twelve clinical reasoning questions given to medical students, four each on three urinary tract infection scenarios: uncomplicated UTI, complicated UTI, and pyelonephritis. They are Key Features Questions, which present a case and ask for the decisions critical to resolving it, and they are marked with partial credit – which is why the scores are fractional rather than right or wrong. THE QUESTIONS THEMSELVES ARE NOT SHOWN: the case vignettes are not published in the data deposit and the article is paywalled, so only the scenario and the question number are given.
girma_2021_oslo3	Item wording is the canonical English OSSS-3 (Kocalevent et al. 2018); the study itself published no item text, and it does not state which language its Ethiopian secondary-school respondents were surveyed in.
girma_2021_phq9	Item text is the canonical English PHQ-9; the study administered the questionnaire in a local Ethiopian language (Amharic or Afaan Oromo) and neither the paper nor its supplement publishes that wording.
gizaw_2023_phq9	Item wording here is the canonical English PHQ-9; the study administered an Amharic translation whose exact wording is published neither in the paper nor in its Stata supplement. The item codes were tied to items by the supplement own variable labels, which are abbreviated symptom mnemonics rather than full item text.
gmc_kay_2025	For gmc_xxx_03_r and gmc_xxx_04_r the stored resp is already reverse-scored by the study’s authors, so resp 1 means the respondent chose Strongly agree and resp 7 Strongly disagree; the option text is written accordingly.
gobbens_2018_adl	The item text here is not the wording respondents read: the study was administered in another language and only an English version of the instrument could be recovered
gobbens_2018_iadl	The GARS was administered in Dutch to this sample of Dutch adults aged 75 and older, but no Dutch wording is published in the paper or its SPSS supplement, so the item and option text shipped here is the authors’ own English from the deposited file’s variable and value labels rather than the sentences respondents read.
goksel_2026_embarrassment_appeasement	‘[name]’ stands for the target person’s name, which the survey filled in for each participant (three names varied across participants); the names used are not recorded here.
goksel_2026_embarrassment_emb_perception	‘[name]’ stands for the target person’s name, which the survey filled in for each participant; the names used are not recorded here.
goksel_2026_embarrassment_moralityandwarmth	‘[name]’ stands for the target person’s name, which the survey filled in for each participant; the names used are not recorded here.
goksel_2026_embarrassment_puremorality	‘[name]’ stands for the target person’s name, which the survey filled in for each participant; the names used are not recorded here.
goksel_2026_embarrassment_purewarmth	‘[name]’ stands for the target person’s name, which the survey filled in for each participant; the names used are not recorded here.
goksel_2026_embarrassment_warmth_competence	‘[name]’ stands for the target person’s name, which the survey filled in for each participant; the names used are not recorded here.
goksel_2026_embarrassment_wcm	The instructions name the rated person as [name] (Study 2); Study 6 respondents read the piped actor name in that position and Study 4 respondents read ‘this person’, and the three studies presented the 12 adjectives in different orders.
goldberg_2018_bri_bri	The response-option wording shipped here is the scale definition printed on page 1 of the questionnaire. The running header on the later pages abbreviates the top two anchors differently – “THREE to 15 times in the past year” and “15 or MORE times in the past year” – which places exactly 15 occurrences in category 5, whereas the page-1 wording places it in category 4. The SPSS value labels agree with the abbreviated header. The page-1 definitional wording was shipped; the boundary case of exactly 15 times is ambiguous in the source form itself.
goldberg_2018_bri_feel	The four positively worded items (feel5, feel10, feel14, feel17) are stored as administered and are NOT reverse-scored: their means are 3.95-4.32 while every other item averages 1.09-2.46, so a depression total must reverse them first.
goldberg_2018_bri_food	In this table resp=5 is ‘Not Applicable’ - the questionnaire’s explicit escape for items that do not apply (e.g. ‘take the skin off chicken’ for a non-chicken-eater) - so the scale is ordinal only over 1-4 (Usually or Always to Rarely or Never) and 5 should be treated as a separate category rather than the low end. It is used at 0.1-11.1% per item.
goldberg_2018_bri_ofood	This table pools two response scales under one resp column: items ofood1-ofood18 (page 14 of the form) are anchored 1 = ‘Never or less than once a MONTH’ to 5 = ‘5 times a WEEK or more’, while ofood19-ofood29 (page 15) are anchored 1 = ‘Never or less than once a WEEK’ to 5 = ‘About once a DAY or more’. The codes 1-5 are ordinal within each block but are not on a common frequency metric across the two, so resp values must not be compared or summed across the ofood1-18 and ofood19-29 blocks without rescaling. The shipped option_text gives the correct anchor per item.
goldberg_2018_dop_avocational_interests	The questionnaire prints the activity ‘Wrote poetry’ twice (ai_82 and ai_199). The two codes are separate responses to identical wording (r = 0.90), not a transcription error.
goldberg_2018_eps_adjectives	This table also contains five non-adjective items from the same questionnaire on different response scales (expo911 and famex911 no/yes, rhealth and rhltyr 1-5, rpain 1-6); only the other 237 items use the 1-7 adjective rating scale.
goldberg_2018_eps_s11	The instructions break off mid-sentence (‘…statements are about’) exactly as they do in the printed questionnaire; nothing was lost in extraction.
goldberg_2018_ipip	This table pools three response formats under resp 1-5: most items use the IPIP accuracy scale, but n1-n24 are the Behavioral Report Inventory’s past-week frequency items (the same responses as goldberg_2018_bri_feel) and b1-b31 are Curious Experiences Survey frequency items (the same responses as BRI curio1-curio31). Response options are blank for 1,286 items (the h, x, e and s forms), whose anchors no deposit publishes. Item v30 was administered as “Have created something of beauty in the last year”, the reverse of the wording on the IPIP website.
goldberg_2018_sbo	resp=3 on this table is the form’s middle box, which is printed as ‘?’ rather than a number and is labelled ‘Neither agree nor disagree: neutral, uncertain, or don’t understand the statement’ – so a 3 combines genuine neutrality with uncertainty and non-comprehension, and is not purely a scale midpoint
goldberg_2018_sdv_adjectives	The midpoint of this 1-7 scale was printed as “?” and labelled “Uncertain, Neutral, or Don’t know word”, so resp = 4 mixes neutral self-ratings with respondents who did not know the adjective.
goldberg_2018_sdv_ipip_temperament	Item text follows the printed form respondents answered, which differs from the deposit’s item labels (and the IPIP wording) on four items; the substantive one is d95, printed as “Find it easy to tune out background noises when trying to concentrate.” where the label adds “on something else”.
goldberg_2018_spa_skill_proficiency	These are ranks, not ratings: each respondent was asked to rank the nine skills from 1 (their best) to 9 (their worst), using each number once, so lower values mean a stronger self-rated skill; 172 of the 700 respondents who answered all nine repeated at least one rank.
gomez_2022_qcae	Two things to know before using this table. (1) The four reverse-worded items – QCAE1r, QCAE2r, QCAE17r and QCAE29r – are stored already reverse-scored, so for those four (and only those four) resp 1 means the respondent chose ‘Strongly Agree’ and resp 4 ‘Strongly Disagree’; the response-option text shipped here is flipped accordingly, and no further reversal is needed to score the scale. (2) The source paper’s Methods describes the scale as running 1 (strongly agree) to 4 (strongly disagree), which contradicts the deposited .sav’s own value labels; the labels are correct – summing as stored reproduces the published QCAE norms (Cognitive M = 58.8 vs 57.1 in Powell 2018; Affective M = 34.6 vs 33.8), while the paper’s direction would give 36.2 and 25.4. Wording for the four unlabelled reverse items comes from the canonical QCAE rather than this study’s file; items 2 and 29, both narrative-context items, cannot be told apart from the data and were assigned from the instrument’s documented item sourcing.
gordils_2021_behavioral_avoidance	The item-to-text alignment is inferred from the order items appear in the source, not from an explicit code-to-text mapping
gordils_2021_discrimination	The item-to-text alignment is inferred from the order items appear in the source, not from an explicit code-to-text mapping
gordils_2021_intergroup_anxiety	The item-to-text alignment is inferred from the order items appear in the source, not from an explicit code-to-text mapping
gordils_2021_interracial_comp	The item-to-text alignment is inferred from the order items appear in the source, not from an explicit code-to-text mapping
gordils_2021_interracial_trust	The four items are stored in the TRUST direction (higher = more trust, 1 = ‘Not at all’ to 7 = ‘Completely’), whereas the paper’s own analyses use the reverse-scored composite MISTRUST = mean(8 - TRUST1..4); recomputing that composite from these columns reproduces the published Study 1 effect exactly, t(845) = 3.318 against a published 3.32. The study’s appendix also prints two different anchor descriptions for this scale, and the response labels shipped here are the one the paper’s Methods section states (1 = not at all, 7 = completely).
gpt4mcq_young_2025	The OSF deposit prints two different sets of 20 ChatGPT-4 items under the same codes AIQ1-AIQ20: the labels inside the SPSS data file, and a non-overlapping AI block in Supplementary Materials 2. IRW ships the SPSS labels, because P(resp=1) under them reproduces the paper’s published proportion-correct for all 20 items to within 0.0005, while the Supplementary Materials 2 keying would be chosen by 10% of respondents on average
grandahl_2017_hpv_beliefs	The questionnaire was administered in Swedish; the item text shown is the study’s own English, taken from the variable labels of its SPSS deposit, and no Swedish original is published. Response direction is reversed relative to that file (IRW 1 = Totally disagree, 5 = Totally agree; the .sav codes the reverse), and its sixth level “Do not know” is dropped as missing, which is why per-item n runs from 202 to 712; Q23o_Attend_screening was answered almost only by girls (232 female, 4 male)
graves_2021_brief_cope	The item and response wording is Carver’s published Brief COPE rather than this study’s own form, which was not published (the paper describes the anchors only as 1 ‘doesn’t do at all’ to 4 ‘does a lot’). Codes BC1-BC28 follow Carver’s numbering, which the study’s own subscale scores confirm for all 14 two-item subscales but not for the order within each pair. Carver’s printed instructions are for a presurgery interview and were not what these students read, so no instructions are included.
guatemala_2024_homes_avenues	The English item, instruction and response-option wording for this table was produced by IRW, not by INE Guatemala, which publishes the ENCASBA 2024 questionnaire in Spanish only; the administered Spanish is in the base fields. Note also: on a_p04a06a1-a8, resp 3 is NS (don’t know), kept as a response category, so those items are yes/no/don’t-know rather than ordinal; on the 1-10 satisfaction item a_p04a06b only the endpoints have option text, and 387 ‘no sabe / no responde’ answers remain as rows with resp NULL.
guatemala_2024_homes_country	The English item, instruction and response-option wording for this table was produced by IRW, not by INE Guatemala, which publishes the ENCASBA 2024 questionnaire in Spanish only; the administered Spanish is in the base fields. Note also: on cp_p09a01-cp_p09a04 INE codes the middle category (Regular / Igual) as 5, after the four ordered options, so resp 1-5 is not ordinal as stored; don’t-know and no-answer responses remain as rows with resp NULL (6,424 of 96,404 rows).
guatemala_2024_homes_drainage	The English item, instruction and response-option wording for this table was produced by IRW, not by INE Guatemala, which publishes the ENCASBA 2024 questionnaire in Spanish only; the administered Spanish is in the base fields. Note also: d_p04a02 is a screener (connected to drainage, yes/no); the 2,633 households not connected skipped every other item, and their rows remain with resp NULL, as do 140 ‘no sabe / no responde’ answers on d_p04a02b; on d_p04a02a1-a5, resp 3 is NS (don’t know), kept as a response category; on the 1-10 item d_p04a02b only the endpoints have option text.
guatemala_2024_homes_emergencies	The English item, instruction and response-option wording for this table was produced by IRW, not by INE Guatemala, which publishes the ENCASBA 2024 questionnaire in Spanish only; the administered Spanish is in the base fields. Note also: on u_p04a10a1-a7, resp 3 is NS (don’t know), kept as a response category, and it is common here (up to 1,212 of 6,886 on one item); on the 1-10 item u_p04a10b only the endpoints have option text, and 942 ‘no sabe / no responde’ answers remain as rows with resp NULL.
guatemala_2024_homes_energy	The English item, instruction and response-option wording for this table was produced by IRW, not by INE Guatemala, which publishes the ENCASBA 2024 questionnaire in Spanish only; the administered Spanish is in the base fields. Note also: e_p04a08 is a screener (connected to an electric grid, yes/no); the 525 households not connected skipped every other item, and their rows remain with resp NULL, as do 217 ‘no sabe / no responde’ answers on e_p04a08b; on e_p04a08a1-a4, resp 3 is NS (don’t know), kept as a response category; on the 1-10 item e_p04a08b only the endpoints have option text.
guatemala_2024_homes_lighting	The English item, instruction and response-option wording for this table was produced by IRW, not by INE Guatemala, which publishes the ENCASBA 2024 questionnaire in Spanish only; the administered Spanish is in the base fields. Note also: l_p04a03 is a screener (the neighbourhood has public lighting, yes/no); the 1,706 households without it skipped every other item, and their rows remain with resp NULL, as do 164 ‘no sabe / no responde’ answers on l_p04a03b; on l_p04a03a1-a4, resp 3 is NS (don’t know), kept as a response category; on the 1-10 item l_p04a03b only the endpoints have option text.
guatemala_2024_homes_nutrition	The English item, instruction and response-option wording for this table was produced by IRW, not by INE Guatemala, which publishes the ENCASBA 2024 questionnaire in Spanish only; the administered Spanish is in the base fields. Note also: only households where a member attended a public school in 2024 were asked, and the screener for that is in guatemala_2024_homes_publiceducation (pe_p05a01), not here; the other 4,263 households’ rows remain with resp NULL (62% of the table), as do 125 ‘no sabe / no responde’ answers on sn_p05a01d; on sn_p05a01c1-c4, resp 3 is NS (don’t know), kept as a response category; on the 1-10 item sn_p05a01d only the endpoints have option text.
guatemala_2024_homes_parks	The English item, instruction and response-option wording for this table was produced by IRW, not by INE Guatemala, which publishes the ENCASBA 2024 questionnaire in Spanish only; the administered Spanish is in the base fields. Note also: p_p04a04 is a screener (parks or gardens in the neighbourhood, yes/no); the 5,750 households without them skipped every other item, and their rows remain with resp NULL (73% of the table, leaving 1,136 valid responses per follow-up item), as do 26 ‘no sabe / no responde’ answers on p_p04a04b; on p_p04a04a1-a6, resp 3 is NS (don’t know), kept as a response category; on the 1-10 item p_p04a04b only the endpoints have option text.
guatemala_2024_homes_publicclinic	The English item, instruction and response-option wording for this table was produced by IRW, not by INE Guatemala, which publishes the ENCASBA 2024 questionnaire in Spanish only; the administered Spanish is in the base fields. Note also: despite the table name, these are the questionnaire’s public-hospital items (P05A03); public health centres are in guatemala_2024_homes_publichealth. pc_p05a03 is a screener (a household member attended a public hospital in 2024); the 5,083 households who did not skipped every other item, and their rows remain with resp NULL, as do 42 ‘no sabe / no responde’ answers on pc_p05a03b; on pc_p05a03a1-a13, resp 3 is NS (don’t know), kept as a response category; on the 1-10 item pc_p05a03b only the endpoints have option text.
guatemala_2024_homes_publiceducation	The English item, instruction and response-option wording for this table was produced by IRW, not by INE Guatemala, which publishes the ENCASBA 2024 questionnaire in Spanish only; the administered Spanish is in the base fields. Note also: pe_p05a01 is a screener (a household member attended a public education centre in 2024); the 4,263 households who did not skipped every other item, and their rows remain with resp NULL, as do 111 ‘no sabe / no responde’ answers on pe_p05a01b; on pe_p05a01a1-a11, resp 3 is NS (don’t know), kept as a response category; on the 1-10 item pe_p05a01b only the endpoints have option text.
guatemala_2024_homes_publichealth	The English item, instruction and response-option wording for this table was produced by IRW, not by INE Guatemala, which publishes the ENCASBA 2024 questionnaire in Spanish only; the administered Spanish is in the base fields. Note also: ph_p05a02 is a screener (a household member attended a public health centre in 2024); the 3,953 households who did not skipped every other item, and their rows remain with resp NULL, as do 84 ‘no sabe / no responde’ answers on ph_p05a02b; on ph_p05a02a1-a12, resp 3 is NS (don’t know), kept as a response category; A8, A9 and A12 are negatively worded, so Sí is the unfavourable answer there; on the 1-10 item ph_p05a02b only the endpoints have option text.
guatemala_2024_homes_roads	The English item, instruction and response-option wording for this table was produced by IRW, not by INE Guatemala, which publishes the ENCASBA 2024 questionnaire in Spanish only; the administered Spanish is in the base fields. Note also: on r_p04a07a1-a4, resp 3 is NS (don’t know), kept as a response category; A4 (asked for a bribe) is negatively worded, so Sí is the unfavourable answer there; on the 1-10 item r_p04a07b only the endpoints have option text, and 490 ‘no sabe / no responde’ answers remain as rows with resp NULL.
guatemala_2024_homes_safety	The English item, instruction and response-option wording for this table was produced by IRW, not by INE Guatemala, which publishes the ENCASBA 2024 questionnaire in Spanish only; the administered Spanish is in the base fields. Note also: on s_p04a09a1-a7, resp 3 is NS (don’t know), kept as a response category, and it is common here (up to 1,312 of 6,886 on one item); A7 (asked for a bribe) is negatively worded, so Sí is the unfavourable answer there; items A1, A2 and A5 speak of emergencies inside this police block, as printed; on the 1-10 item s_p04a09b only the endpoints have option text, and 627 ‘no sabe / no responde’ answers remain as rows with resp NULL.
guatemala_2024_homes_transparency	The English item, instruction and response-option wording for this table was produced by IRW, not by INE Guatemala, which publishes the ENCASBA 2024 questionnaire in Spanish only; the administered Spanish is in the base fields. Note also: the items use different scales (tc_p10a01a1-a12a 1-4 honourable to corrupt, tc_p10a02 1-3, six yes/no items, tc_p10b05 1-5 where 1 is most frequent); don’t-know and no-answer answers remain as rows with resp NULL (35% of the table); tc_p10b01a and tc_p10b02a were asked only of respondents who said Sí to tc_p10b01 and tc_p10b02; tc_p10a01a12a rates an ‘other’ institution each respondent named, whose name is not recorded, and only 27 households answered it.
guatemala_2024_homes_transport	The English item, instruction and response-option wording for this table was produced by IRW, not by INE Guatemala, which publishes the ENCASBA 2024 questionnaire in Spanish only; the administered Spanish is in the base fields. Note also: the block was asked only of households that used the municipal collective (public) bus in 2024 (441 of 6,886), so the 6,445 households using private collectives or no public transport have resp NULL on every item, as do 11 pt_p05a04b answers of ‘no sabe / no responde’; on pt_p05a04a1-a9, resp 3 is NS (don’t know), kept as a response category, so those items are yes/no/don’t-know rather than ordinal.
guatemala_2024_homes_trash	The English item, instruction and response-option wording for this table was produced by IRW, not by INE Guatemala, which publishes the ENCASBA 2024 questionnaire in Spanish only; the administered Spanish is in the base fields. Note also: the block was asked only of households that dispose of most of their garbage through the municipal service (2,202 of 6,886), so the 4,684 households using a private service or another method have resp NULL on every item, as do 86 t_p04a05b answers of ‘no sabe / no responde’; on t_p04a05a1-a4, resp 3 is NS (don’t know), kept as a response category, so those items are yes/no/don’t-know rather than ordinal.
guatemala_2024_homes_water	The English item, instruction and response-option wording for this table was produced by IRW, not by INE Guatemala, which publishes the ENCASBA 2024 questionnaire in Spanish only; the administered Spanish is in the base fields. Note also: the block was asked only of households whose drinking water comes from a municipal or community service (5,487 of 6,886), so the 1,399 households with a private well, no service or another source have resp NULL on every item, as do 245 w_p04a01b answers of ‘no sabe / no responde’; on w_p04a01a1-a5, resp 3 is NS (don’t know), kept as a response category, so those items are yes/no/don’t-know rather than ordinal.
gumus_2025_dietarian_identity	Fourteen of this table’s items are stored reverse-coded relative to the agreement scale, so for DIQ9-DIQ19 and DIQ31-DIQ33 resp=1 means ‘Strongly agree’ and resp=7 ‘Strongly disagree’ (the shipped option_text is per item and reflects this). DIQ19 looks like a depositor error rather than reverse wording: it is a positively-worded prosocial-motivation item, correlates -0.72 to -0.80 with the other five items of its own subscale as stored, and the paper’s published Prosocial Motivation statistics (mean 3.85, SD 2.02, alpha 0.95) reproduce only if DIQ19 is un-reversed (as stored the block gives 3.97 / 1.39 / 0.68).
guo_2025_ar	Respondents read a Chinese version of these items; the wording shown is the study authors’ own English rendering from their data deposit, as the Chinese wording was not published.
guo_2025_ase	Respondents read a Chinese version of these items; the wording shown is the study authors’ own English rendering from their data deposit, as the Chinese wording was not published. The four negatively worded items (ase_14, ase_16, ase_17, ase_20) correlate positively (pairwise +0.25 to +0.59) with the positively worded items and the paper’s reported scale mean reproduces only with them as stored, so it is unclear whether their stored values are raw agreement or already reverse-scored.
gwen_2025_interpersonal_comm	The English translations (item_text_translated) were written by IRW from the Indonesian statements respondents read; they are not the study’s own.
gwen_2025_mental_health	As for gwen_2025_interpersonal_comm: IRW-written English from the Indonesian statements.
gwen_2025_self_esteem	As for gwen_2025_interpersonal_comm: IRW-written English from the Indonesian statements.
gwen_2025_social_support	As for gwen_2025_interpersonal_comm: IRW-written English from the Indonesian statements.
ha_2026_aspire_affect	Item text is published for 5 of the 7 affect items (happy, energetic, low, irritable, distressed); insec and attr have no wording in the source paper or deposit, so their item_text and anchors are blank, and only the 1 (Not at all) and 7 (Very much) anchors of the 1-7 scale are labelled.
habibi_2021_meim	The study’s own Persian questionnaire (supplement S3) prints five response columns (کاملا مخالفم / مخالفم / نظری ندارم / موافقم / کاملا موافقم) while the paper describes, and the data store, a four-point 1-4 scale, so only the two endpoints the paper states are labelled here and option_text for resp 2 and 3 is left blank; the English item wording is the paper’s own Table 2, but the English instructions are IRW’s translation of the Persian form
han_2015_peer_assisted_learning	The item text here is the study’s own English questionnaire (S1 Text, Section B, items 1-12) verbatim, with the anchors taken from that questionnaire’s own column headers over its printed 1-5 codes. What the source never states is the administration language: the study was run at a Korean medical school, but neither the paper nor its supplements publish a Korean version, and the English S1 Text is the only questionnaire in the deposit. So the shipped English may be a parallel version rather than what students actually read, and the table deliberately carries no language or *_translated columns rather than assert a Korean original that nobody published.
han_2026_gad7	In han_2026_gad7 only GAD01-GAD06 are GAD-7 items, and GAD02/GAD03 are the study’s own order (excessive worry, then uncontrollable worry) rather than the canonical numbering used in the paper’s Table 2; GAD07 is not a GAD-7 item at all but the survey’s loss-of-interest question, answered No/Cannot judge/Yes and stored as 0/1/2, so its response scale differs from the other six items’ 0-3 frequency scale.
han_2026_isi	Response codes are not uniform across this table’s items: ISI01 and ISI07 run 0-4 while ISI02-ISI06 run 1-5, because the deposited file shifted only two of the seven columns down by one. All seven are the same five-point scale, and the shipped option text is mapped to each item’s own codes (lowest anchor at resp=0 for ISI01/ISI07, at resp=1 for the rest); a total score summed from the raw codes is inflated by 5. Note also that the paper’s Table 2 labels ISI05/ISI06/ISI07 as noticeability/distress/interference, whereas the study’s own analysis script maps those columns to the daily-life-impact, quality-of-life-prominence and worry questions – the item text here follows the script and the raw file’s column headers
han_2026_phq9	Item wording for this table is the Chinese the study administered, with the authors’ own English alongside it, but the response-option labels ship in English only: the deposit publishes the anchors (Never / Several days / More than half the days / Nearly every day) in English and never in Chinese.
hannachi_2025_eco_anxiety_cas	Responses on this table are 0-4 on an agreement-type scale that the study documents only in English – its supplementary materials state “Response categories: 0 = not at all to 4 = completely” – and NOT on the Climate Change Anxiety Scale’s own published 1-5 frequency anchors (1 = Never to 5 = Almost always); the administered French anchor wording was never published, so option_text is left blank rather than guessed
hao_2025_emotional_regulation	Despite the table name, these four items are the ego-resilience (ER) factor of a foreign-language resilience scale, not an emotional-regulation measure. The study says it adapted the scale from Guo and Li (2022) but did not publish its own wording, so the Chinese item text and its English are Guo and Li’s published Foreign Language Learning Resilience Scale (items 1, 2, 3 and 6), and what these respondents read may differ; ER1-ER4 are matched to those items by order. The respondents were 650 junior high school students aged 12-15. The study reports a 7-point scale (1 = Strongly disagree, 7 = Strongly agree) ‘standardized to a 5-point scale’, but the data hold 1-5 and how the labels map onto those values is not published, so the response options are unlabelled.
hao_2025_metacognitive_regulation	These items are the metacognitive-resilience factor of a foreign-language resilience scale. The study says it adapted the scale from Guo and Li (2022) but did not publish its own wording, so the Chinese item text and its English are Guo and Li’s published Foreign Language Learning Resilience Scale (items 9-15), and what these junior high school respondents read may differ; MR1-MR7 are matched to those items by order, which item means from Guo and Li’s sample support but do not confirm. Guo and Li’s English and Chinese differ on MR7 (‘I am determined when I study a foreign language’ against Chinese meaning ‘when I decide to learn a foreign language I will find ways to persist’). The study reports a 7-point scale but the data hold 1-5, so the response options are unlabelled.
hao_2025_social_regulation	These items are the social-resilience factor of a foreign-language resilience scale. The study says it adapted the scale from Guo and Li (2022) but did not publish its own wording, so the Chinese item text and its English are Guo and Li’s published Foreign Language Learning Resilience Scale (items 17-24), and what these junior high school respondents read may differ; SR1-SR8 are matched to those items by order. Guo and Li’s English and Chinese differ on SR1 (‘seek encouragement from my classmates’ against Chinese ‘seek help’), and their English typos in SR7 and SR8 are kept as printed. The study reports a 7-point scale but the data hold 1-5, so the response options are unlabelled.
hayek_2022_attitude	The two negatively worded items (att3, att4) are shipped with reversed response anchors (resp = 2 means ‘Strongly disagree’), following the paper’s statement that they were reverse coded; the data file carries no value labels for these items, and att3’s negative correlations with the two positively worded items leave that direction less than certain for att3.
hayek_2022_self_efficacy	The item-to-text alignment is inferred from the order items appear in the source, not from an explicit code-to-text mapping
hayek_2022_subj_norm	The three items are printed in the paper’s Measures section as a single slash-joined sentence (“My father/my mother/my teacher expects me to get good academic grades”), and are shipped here expanded into three separate sentences – the one deviation from literal transcription in this table. Which code refers to which referent is not a guess: the deposited .sav labels S_N_1/2/3 as father/mother/teacher and the processing script renames that explicit column list number-preservingly. The response anchors are the .sav’s own value labels (-2 Strongly disagree .. 2 Strongly agree) and match the live responses exactly. The study ran in Lebanese secondary schools but never states the administration language, and the deposit is English-only, so no language or *_translated columns are shipped rather than assert an original nobody published; instructions are blank for the same reason.
he_2019_flipped_classroom_attitudes	The cov_group labels in this response table are swapped: respondents labelled lecture_based (n=81) are the flipped-classroom group and those labelled flipped_classroom (n=56) are the lecture-based group, per the paper’s group sizes and Table 6 means. Item wording is the authors’ English rendering of a questionnaire presumably administered in Chinese.
he_2019_flipped_classroom_satisfaction	Item wording is the English published in the article; the questionnaire was presumably administered in Chinese and no Chinese version was published. The response table’s cov_group labels are swapped: rows labelled flipped_classroom are the lecture-based arm (n=56) and vice versa (n=81).
heekerens2025_dss	The item text here is the administered German, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. option_text is already English, so its translated column stays empty.
heekerens2025_phq	The study’s German PHQ-8 form was not published, so the item text is the standard German PHQ-D wording, with the official English PHQ-9 (items 1-8) alongside; that the study used this wording is inferred from its description as ‘the German version of the PHQ-8’. Item order is assumed to follow the standard PHQ numbering. Responses are stored 1-4, the usual 0-3 scores shifted up by one (1 = Überhaupt nicht / Not at all, 4 = Beinahe jeden Tag / Nearly every day).
heekerens2025_sdq	The item text here is the administered German, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. This table is labelled the Somatoform Dissociation Questionnaire (SDQ-20), but its 20 German item texts are identical to heekerens2025_dss’s, differing only in one typo. The SDQ-20 is a different instrument, so one of the two tables is almost certainly carrying the other’s item text. The published text is translated as it stands and is not repaired here; the discrepancy needs resolving at source.
hellstrom_2019_isi	The ISI was administered in Swedish, but no Swedish wording is published in the deposit or the paper, so the item and option text shown here is the study’s own English labelling; the item text is the deposit’s short domain labels (e.g. “Quality of life”) rather than the full published ISI question stems.
herreromontes_2022_audit	Respondents completed the Spanish version of the AUDIT; the item text here is the study’s own English codebook wording, close to but not identical with WHO’s English original. The AUDIT is a World Health Organization instrument (Babor et al. 2001, WHO/MSD/MSB/01.6a) and WHO reserves all rights in it; WHO permits free reproduction for non-commercial purposes, which is the basis on which it appears here.
hewei_2022_msva_purchase	The questionnaire was administered in China (Questionnaire Star platform) and so almost certainly in Chinese, but neither the paper nor its single supplement publishes any Chinese wording, so the item text shown is the study’s own English. Only the two endpoints of the 5-point scale are labelled in the source (1 = strong disagreement, 5 = strong agreement); points 2-4 are left blank.
hexaco_ashton_2014_a	These are the IPIP-HEXACO scales (Ashton, Lee and Goldberg), public-domain items from ipip.ori.org, not the HEXACO-PI-R the table description names. Item text is the dataset codebook’s wording, which prefixes every IPIP statement with ‘I’. Response option 5 is shipped as ‘slightly agree’; the codebook prints it as ‘slightly disagree’ (a duplicate of option 3), which we read as a typo. The original test page is gone, so the label respondents saw could not be checked. The prefix makes item AFlex3 read ‘I when interacting with a group of people, am often bothered by at least one of them.’
hexaco_ashton_2014_c	These are the IPIP-HEXACO scales (Ashton, Lee and Goldberg), public-domain items from ipip.ori.org, not the HEXACO-PI-R the table description names. Item text is the dataset codebook’s wording, which prefixes every IPIP statement with ‘I’. Response option 5 is shipped as ‘slightly agree’; the codebook prints it as ‘slightly disagree’ (a duplicate of option 3), which we read as a typo. The original test page is gone, so the label respondents saw could not be checked.
hexaco_ashton_2014_e	These are the IPIP-HEXACO scales (Ashton, Lee and Goldberg), public-domain items from ipip.ori.org, not the HEXACO-PI-R the table description names. Item text is the dataset codebook’s wording, which prefixes every IPIP statement with ‘I’. Response option 5 is shipped as ‘slightly agree’; the codebook prints it as ‘slightly disagree’ (a duplicate of option 3), which we read as a typo. The original test page is gone, so the label respondents saw could not be checked.
hexaco_ashton_2014_h	These are the IPIP-HEXACO scales (Ashton, Lee and Goldberg), public-domain items from ipip.ori.org, not the HEXACO-PI-R the table description names. Item text is the dataset codebook’s wording, which prefixes every IPIP statement with ‘I’. Response option 5 is shipped as ‘slightly agree’; the codebook prints it as ‘slightly disagree’ (a duplicate of option 3), which we read as a typo. The original test page is gone, so the label respondents saw could not be checked.
hexaco_ashton_2014_o	These are the IPIP-HEXACO scales (Ashton, Lee and Goldberg), public-domain items from ipip.ori.org, not the HEXACO-PI-R the table description names. Item text is the dataset codebook’s wording, which prefixes every IPIP statement with ‘I’. Response option 5 is shipped as ‘slightly agree’; the codebook prints it as ‘slightly disagree’ (a duplicate of option 3), which we read as a typo. The original test page is gone, so the label respondents saw could not be checked.
hexaco_ashton_2014_x	These are the IPIP-HEXACO scales (Ashton, Lee and Goldberg), public-domain items from ipip.ori.org, not the HEXACO-PI-R the table description names. Item text is the dataset codebook’s wording, which prefixes every IPIP statement with ‘I’. Response option 5 is shipped as ‘slightly agree’; the codebook prints it as ‘slightly disagree’ (a duplicate of option 3), which we read as a typo. The original test page is gone, so the label respondents saw could not be checked.
hicks_2020_bioveda	The BioVEDA items are multiple choice, but the IRW resp column is a 0/1 correctness score rather than the option chosen, so option_text records Incorrect/Correct rather than the four answer alternatives (which the study does publish, with its key, in its S1 File). Ten of the sixteen items also depend on a figure, equation or data table that item_text cannot carry.
himmelstein-impossible_question-2025	18 of the 90 items (the IQ_* codes) are deliberately unanswerable – they ask about invented battles, shapes and places, and the correct response is to select “Opt-out”. So resp = 1 means the respondent either picked the keyed answer or correctly opted out, and resp = 0 covers both a wrong answer and opting out of a question that did have one; note that opting out of an answerable item is scored incorrect even though the instructions told participants it would not be scored. option_text is empty by design: each trial offered two alternatives plus “Opt-out”, but resp is a 0/1 correctness score rather than which option was chosen, so the option texts cannot be mapped onto it – the correct option is given in correct_response instead
hirwa_2024_antibiotic_attitudes	Respondents answered in Kinyarwanda; the item text shipped here is the study’s own English questionnaire, as the Kinyarwanda wording is not published in the deposit. The five-point Likert scale described in the paper (strongly agree to strongly disagree) is collapsed to three levels in the deposited data, so resp 0/1/2 read as Disagree/Neutral/Agree for every item – including the two reverse-worded items, which are stored as raw agreement rather than on the paper’s correct/neutral/incorrect scoring rubric.
hoorani_2022_child_help	The item wording shown here is the English questionnaire wording printed in the paper’s Table 1; the Young Lives India older-cohort survey was administered in the child’s own language (Telugu in the Andhra Pradesh sample) and no non-English wording is published in the paper or its Stata supplement.
hoorani_2022_epreas	The item wording is the English printed in the paper; the Young Lives India caregiver questionnaire was administered in the respondent’s own language (Telugu in the Andhra Pradesh sample), and these round-two caregiver answers are repeated unchanged as the table’s wave 2.
hoorani_2022_ps	The wording shown is the English from the study’s Stata supplement and paper; the Young Lives India older-cohort survey was administered in the child’s own language (Telugu in the Andhra Pradesh sample), and no non-English wording is published. Items 2-4 are negatively worded and stored raw.
hoorani_2022_sp	The four self-pride items are shipped in English, taken from the study’s own Stata deposit, but neither the paper nor the deposit states what language they were administered in: this is the Young Lives older cohort in Andhra Pradesh, India, and the paper’s only remark on language concerns a different measure (“conducted in the native language or the language that the adolescent chose”). A non-English administration cannot be ruled out, so the table deliberately carries no language or *_translated columns rather than assert an original nobody published. Two further things a reader should know: no instruction or block-prompt text exists in either source, so those fields are NA rather than invented; and SP1-SP4 were administered in one round only (Young Lives Round 3, 2009 – IRW wave 2), which is why every item has 535 responses and wave 1 has none.
hori_2019_radiation_risk	The response codes in this table are the authors’ analysis recoding rather than the raw answers, so three items carry no option text at all: Q1 (age), Q6 (years living in the area) and Q8 (years employed) were administered as open write-in numbers and are stored as a two-level split (resp=2 is ‘older than 40 years’, ‘more than 7 years’ and ‘more than 21 years’ respectively, per the paper’s Tables 1-2), and six further items were collapsed from their administered scales – Q5 from six distance bands and Q13, Q14, Q18, Q20 and Q21 from 4-point anchors – with the option text for those showing the administered anchors joined by ’ / ’ on the assumption of an even 2+2 split, which the paper does not state.
horiuchi_2024_attachment	Only 13 of the 20 items carry wording: the paper’s item list is not in the questionnaire’s column order (it cannot even be reconciled at the set level with the 13 items the paper reports retaining, and it prints one wording twice), so the 7 items dropped before any per-item statistic was published cannot be tied to a code and ship with item_text blank. The 13 that do carry wording were matched through the paper’s factor table and its Survey 2 readministration, which fixes them as a group but leaves 11 of 78 possible pairwise swaps among them unexcluded. The instrument was administered in Japanese and no Japanese wording is published anywhere in the article or its supplements, so the English shown is the study’s own; the response anchors are the raw file’s marks (x=0, triangle=1, circle=2), the only coding the study published.
horiuchi_2024_dissociation	The item wording shipped for this table is English, while the checklist was administered in Japanese to teachers and orphanage staff; no Japanese wording is published anywhere in the study’s deposit, so the English the authors printed in their own questionnaire supplement is shipped in its place. The response options are a further step removed: the study published no anchors and its raw data file records only the marks ×, △ and 〇 keyed to 0, 1 and 2, so the 0 = NOT TRUE / 1 = SOMEWHAT or SOMETIMES TRUE / 2 = VERY TRUE labels come from Putnam’s published CDC form rather than from anything these raters saw.
horiuchi_2024_rsmsm	The item text here is not the wording respondents read: the study was administered in another language and only an English version of the instrument could be recovered; the item-to-text alignment is inferred from the order items appear in the source, not from an explicit code-to-text mapping
hua_2023_efl_course_experience	The item and option wording shown is the study’s own Chinese as administered. The English in the *_translated columns was produced by IRW, not by the study’s authors, and has not been reviewed or back-translated by a Chinese speaker – treat it as a reading aid rather than a validated English instrument. The Chinese itself is the deposit’s own variable and value labels, so the item-to-text mapping needs no inference.
huang_2016_cesd	Huang et al. publish no item wording, so the text here is Radloff’s (1977) original English CES-D, transcribed from a later CC BY reprint, mapped to the deposit’s unlabelled CESD_1-CESD_20 columns by canonical item number. 30 of the 114 participants were assessed in French, and that French wording is not published anywhere in the record. Items 4, 8, 12 and 16 are stored already reverse-scored, so their frequency anchors run in reverse (resp 0 = ‘Most or all of the time’) relative to how the positively worded stem reads.
huang_2023_d_scale	Two things to know before using this text. First, the wording is not this study’s: the study publishes no item text of its own, so what is shown is the standard Chinese rendering of Campbell’s Index of Well-Being taken from a third-party reproduction of the scale, and the English is IRW’s own translation of that Chinese rather than the authors’ or the instrument’s published English. Second, the eight affect items are bipolar semantic-differential pairs, so item_text is blank by design and the words sit in the option text for scale points 1 and 7 only. Their assignment to codes d1-d9 follows the instrument’s published item order and is confirmed by the data only up to polarity class – the live correlations reproduce the published “items A, C, F and G are flipped” pattern exactly ({d1,d3,d6,d7,d9} keyed high = good), but the order within each of those groups is not independently established.
huang_2023_medseq	The source describes these items as answered on a 1-6 scale (1 = strongly disagree; 6 = strongly agree), but the deposited data contain only values 1-5 on every item (no 6 among 37,818 responses), so the response options are left unlabelled.
hui_2024_gbfs	The item text here is the administered Chinese from the study’s S1 File. The English in item_text_translated is not a translation of it but the original English General Benefit Finding Scale (Cassidy, McLaughlin & Giles 2014) aligned item for item, so it diverges where the Chinese adaptation did – item b18 reads roughly ‘made me more confident in others’ in Chinese against the original’s ‘Led me to feel more positive about others’. The English instructions and response anchors have no published source and were written by IRW.
hyatt_2023_aggression_aeq	This table pools the whole development of the Aggression Expectancy Questionnaire: prelim1_, prelim2_ and s1_* are the 84-, 40- and 31-item development pools under each study’s own item numbering (only 66 and 27 of the first two pools are in the deposited data), and aeq01-aeq16 are the final form answered in Studies 2 and 3; the five prelim1 items ending _r are stored reverse-scored, so their option_text runs from ‘Very likely’ at 1 to ‘Very unlikely’ at 5; and the paper’s Tables 2-3 list aeq09-aeq16 in a different order (and with ‘The other person will really suffer’ for aeq13) than the final instrument and the Study 3 data file, which this table follows
hyatt_2023_aggression_s3_acme	The 22 items whose codes end in ‘r’ (e.g. acme_2dr, acme_12rr, acme_31cr) are stored already reverse-scored, so for them resp 1 = Strongly agree and 5 = Strongly disagree, the opposite of the source file’s value labels; the option_text shipped here follows the stored direction, and higher resp means more empathy on every item
hypersensitive_narcissism	The table is named for the Hypersensitive Narcissism Scale but holds 22 items from two instruments: the 10-item HSNS (items 1-10) followed by the 12-item Dirty Dozen (items 11-22, in the order psychopathy, narcissism, Machiavellianism – the column order of the source data file, which is not the order its codebook lists them in). Both were administered on the same 1-5 scale in the Open-Source Psychometrics Project’s Dark Triad Personality Test, which labels only 1 (disagree), 3 (neutral) and 5 (agree), so option_text is blank at 2 and 4. The item codes are bare integers assigned by IRW processing; the source’s own codes (HSNS1-10, DDM/DDP/DDN1-4) are not carried in the table. The IRW dictionary entry for this table describes and cites only the HSNS and is being corrected separately.
iandolo_2021_asq	The item wording is the study’s Spanish version, which only the 139 Spanish respondents read; the 85 Italian and 130 Japanese respondents read back-translated versions the deposit does not publish. The English in item_text_translated is Feeney, Noller & Hanrahan’s original ASQ wording, not a translation of the Spanish.
iat_poverty	Twelve possible explanations for poverty, each rated for how important it is. They share the stem ‘There are many poor people in the world because of…’ and were answered from 1 (not at all important) to 5 (extremely important). They come from the Project Implicit demo-website dataset, where they sit alongside the race IAT itself. The third response label is printed ‘Moderately Imporant’ in the source and is left as found.
ibrahim_2015_sf36	Patients completed the questionnaires in Malay or English, but only English SF-36 wording is recoverable, so the item stems, instructions and section prompts are RAND’s published English 36-Item Health Survey 1.0 for a partly Malay-language administration. The deposit’s columns carry no labels, so each code is tied to its wording by the SF-36’s own question numbering. The response options are the study’s own recorded categories, which differ from the published instrument in two places: item 1’s lowest category reads ‘not good’ rather than ‘Poor’, and question 11 was administered on four categories rather than the instrument’s five (there is no ‘Don’t know’).
idas_machado2024	The item text here is the administered Portuguese, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. Two item_text values are not items at all but English scoring notes – reversed scoring of item 27 and reversed scoring of item 64 – which have leaked into the item text column and should be removed at source. Source typos are left verbatim.
IJLS_Eersel_2024_DenialAcceptance	The four Brief COPE denial and acceptance items were administered in Dutch, but the Dutch wording is not published in the DataverseNL deposit or the paper’s supplements, so the item and option text here is the study’s own English rendering from its SPSS labels rather than what respondents read. The item codes BC_2, BC_7, BC_10 and BC_12 are this study’s administered numbering, not the Brief COPE-28’s canonical denial (3, 8) and acceptance (20, 24) item numbers.
IJLS_Eersel_2024_IJLSreaction	The item text is the administered Dutch from the paper’s Appendix, with the authors’ own English in the _translated column, but the response anchors are not what respondents read: no Dutch anchors are published, so option_text shows the English labels from the study’s SPSS value labels (1 = Never to 5 = Always).
IJLS_Eersel_2024_IUS	Intolerance of Uncertainty Scale, 12-item short form, rated 1 = Strongly disagree to 5 = Strongly agree. Administered in Dutch; the English here is the translation the deposit supplies.
IJLS_Eersel_2024_Optimism	Life Orientation Test-Revised, six scored items, rated 1-5. Items 2, 3 and 5 are the pessimism-worded ones and are stored ALREADY REVERSE-SCORED, so for those three a high value means disagreement with the printed statement; their option labels are shipped inverted to match, as the source has them. Administered in Dutch; the English here is the translation the deposit supplies. The LOT-R’s authors note that it is a research instrument and not intended for clinical applications.
IJLS_Eersel_2024_Work	The work centrality, UWES work engagement and commitment items were administered in validated Dutch versions, but the Dutch wording is not published in the deposit or the paper’s supplements, so the text here is the study’s own English rendering from its SPSS labels rather than what respondents read. The article describes work engagement as rated 1 = never to 6 = always, but the SPSS file labels the three UWES items on a 7-point Never..Always scale and the data use all seven levels; the shipped anchors are the file’s. WorkCen_3R is stored already reverse-scored, and its anchors run Strongly agree to Strongly disagree accordingly.
ilearnathome_regalado_2025	Item text for this table is the depositors’ own English rendering; the questionnaire was administered in Spanish to teachers in Chiclayo, Peru, and the deposit does not publish the Spanish wording.
imos_1984	The six problems of the 1984 International Mathematical Olympiad, each marked 0-7. The mathematical notation is rendered in plain text (superscripts as ^, the fraction and floor brackets of problem 5 written inline), so use the official paper if you need the exact typeset statement.
imos_1985	The six problems of the 1985 International Mathematical Olympiad, each marked out of 7, with problems 1-3 sat on the first day and 4-6 on the second. NOTE ON NOTATION: the text is extracted from a PDF, so mathematical layout is flattened – superscripts and fractions are lost, so n squared reads as ‘n2’. Use the official paper if you need the exact statement.
imos_1986	The six problems of the 1986 International Mathematical Olympiad, each marked out of 7, with problems 1-3 sat on the first day and 4-6 on the second. NOTE ON NOTATION: the text is extracted from a PDF, so mathematical layout is flattened – superscripts and fractions are lost, so n squared reads as ‘n2’. Use the official paper if you need the exact statement.
imos_1987	The item text is the organisers’ official English version of the IMO 1987 paper. The IMO is sat in each contestant’s own language, so most of the 237 contestants read a translation of this wording rather than the English shown here. Typeset mathematics is rendered in plain-text notation. The official paper’s own typo in problem 3 (a_1 x_2 where a_2 x_2 is meant) is kept as printed.
imos_1988	The item text is the organisers’ official English version of the IMO 1988 paper. The IMO is sat in each contestant’s own language, so most of the 268 contestants read a translation of this wording rather than the English shown here. Typeset mathematics is rendered in plain-text notation. Problem 4’s missing article (‘Show that set of real numbers x’) is the official paper’s own.
imos_1989	The item text is the organisers’ official English version of the IMO 1989 paper. The IMO is sat in each contestant’s own language, so most of the 291 contestants read a translation of this wording rather than the English shown here. Typeset mathematics is rendered in plain-text notation. Problem 6 keeps the official paper’s undefined m where 2n is meant.
imos_1990	The item text is the organisers’ official English version of the IMO 1990 paper. The IMO is sat in each contestant’s own language, so most of the 308 contestants read a translation of this wording rather than the English shown here. Typeset mathematics is rendered in plain-text notation.
imos_1991	The item text is the organisers’ official English version of the IMO 1991 paper. The IMO is sat in each contestant’s own language, so most of the 312 contestants read a translation of this wording rather than the English shown here. Typeset mathematics is rendered in plain-text notation. The official paper numbers the second day’s problems 1-3 again; problem4-problem6 are those three, matched to the official results’ P4-P6.
imos_1992	The six problems of the 1992 International Mathematical Olympiad, each marked out of 7. Contestants sat problems 1-3 on the first day and 4-6 on the second, with four and a half hours for each paper. The problem statements are taken from the official English paper. NOTE ON NOTATION: the text is extracted from a PDF, so mathematical layout is flattened – superscripts and fractions are lost, so n squared reads as ‘n2’ and one-over-n as ‘1 n’. Use the official PDF if you need the exact statement.
imos_1994	The item text is the organisers’ official English version of the IMO 1994 paper. The IMO is sat in each contestant’s own language, so most of the 385 contestants read a translation of this wording rather than the English shown here. Typeset mathematics is rendered in plain-text notation.
imos_1995	The six problems of the 1995 International Mathematical Olympiad, each marked out of 7. Contestants sat problems 1-3 on the first day and 4-6 on the second, with four and a half hours for each paper. The problem statements are taken from the official English paper. NOTE ON NOTATION: the text is extracted from a PDF, so mathematical layout is flattened – superscripts and fractions are lost, so n squared reads as ‘n2’ and one-over-n as ‘1 n’. Use the official PDF if you need the exact statement.
imos_1996	The item text is the organisers’ official English version of the IMO 1996 paper. The IMO is sat in each contestant’s own language, so most of the 424 contestants read a translation of this wording rather than the English shown here. Typeset mathematics is rendered in plain-text notation. The official paper numbers the second day’s problems 1-3 again; problem4-problem6 are those three, matched to the official results’ P4-P6.
imos_1997	The item text is the organisers’ official English version of the IMO 1997 paper. The IMO is sat in each contestant’s own language, so most of the 460 contestants read a translation of this wording rather than the English shown here. Typeset mathematics is rendered in plain-text notation.
imos_1998	The six problems of the 1998 International Mathematical Olympiad, each marked out of 7. Contestants sat problems 1-3 on the first day and 4-6 on the second, with four and a half hours for each paper. The problem statements are taken from the official English paper. NOTE ON NOTATION: the text is extracted from a PDF, so mathematical layout is flattened – superscripts and fractions are lost, so n squared reads as ‘n2’ and one-over-n as ‘1 n’. Use the official PDF if you need the exact statement.
imos_1999	The item text is the organisers’ official English version of the IMO 1999 paper. The IMO is sat in each contestant’s own language, so most of the 450 contestants read a translation of this wording rather than the English shown here. Typeset mathematics is rendered in plain-text notation. Problem 4 reproduces the official paper’s own erratum ‘n not exceeded 2p’ as printed.
imos_2000	The six problems of the 2000 International Mathematical Olympiad, each marked out of 7. Contestants sat problems 1-3 on the first day and 4-6 on the second, with four and a half hours for each paper. The problem statements are taken from the official English paper. NOTE ON NOTATION: the text is extracted from a PDF, so mathematical layout is flattened – superscripts and fractions are lost, so n squared reads as ‘n2’ and one-over-n as ‘1 n’. Use the official PDF if you need the exact statement.
imos_2001	The item text is the organisers’ official English version of the IMO 2001 paper. The IMO is sat in each contestant’s own language, so most of the 473 contestants read a translation of this wording rather than the English shown here. Typeset mathematics is rendered in plain-text notation.
imos_2002	The six problems of the 2002 International Mathematical Olympiad, each marked out of 7. Contestants sat problems 1-3 on the first day and 4-6 on the second, with four and a half hours for each paper. The problem statements are taken from the official English paper. NOTE ON NOTATION: the text is extracted from a PDF, so mathematical layout is flattened – superscripts and fractions are lost, so n squared reads as ‘n2’ and one-over-n as ‘1 n’. Use the official PDF if you need the exact statement.
imos_2003	The six problems of the 2003 International Mathematical Olympiad, each marked out of 7. Contestants sat problems 1-3 on the first day and 4-6 on the second, with four and a half hours for each paper. The problem statements are taken from the official English paper. NOTE ON NOTATION: the text is extracted from a PDF, so mathematical layout is flattened – superscripts and fractions are lost, so n squared reads as ‘n2’ and one-over-n as ‘1 n’. Use the official PDF if you need the exact statement.
imos_2004	The item text is the organisers’ official English version of the IMO 2004 paper. The IMO is sat in each contestant’s own language, so most of the 486 contestants read a translation of this wording rather than the English shown here. Typeset mathematics is rendered in plain-text notation. Problem 3 depends on a picture of the ‘hook’ hexomino, so its item text carries a bracketed description of the figure written by IRW in place of the image; the paper’s own typo ‘rectagle’ is kept as printed.
imos_2005	The item text is the organisers’ official English version of the IMO 2005 paper. The IMO is sat in each contestant’s own language, so most of the 513 contestants read a translation of this wording rather than the English shown here. Typeset mathematics is rendered in plain-text notation. Problem 5’s ‘lie of the sides’ (for ‘lie on’) is the official paper’s own wording, kept as printed.
imos_2006	The item text is the organisers’ official English version of the IMO 2006 paper. The IMO is sat in each contestant’s own language, so most of the 498 contestants read a translation of this wording rather than the English shown here. Typeset mathematics is rendered in plain-text notation.
imos_2007	The six problems of the 2007 International Mathematical Olympiad, each marked out of 7. Contestants sat problems 1-3 on the first day and 4-6 on the second, with four and a half hours for each paper. The problem statements are taken from the official English paper. NOTE ON NOTATION: the text is extracted from a PDF, so mathematical layout is flattened – superscripts and fractions are lost, so n squared reads as ‘n2’ and one-over-n as ‘1 n’. Use the official PDF if you need the exact statement.
imos_2008	The item text is the organisers’ official English version of the IMO 2008 paper. The IMO is sat in each contestant’s own language, so most of the 535 contestants read a translation of this wording rather than the English shown here. Typeset mathematics is rendered in plain-text notation.
imos_2009	The item text is the organisers’ official English version of the IMO 2009 paper. The IMO is sat in each contestant’s own language, so most of the 565 contestants read a translation of this wording rather than the English shown here. Typeset mathematics is rendered in plain-text notation.
imos_2010	The item text is the organisers’ official English version of the IMO 2010 paper. The IMO is sat in each contestant’s own language, so most of the 516 contestants read a translation of this wording rather than the English shown here. Typeset mathematics is rendered in plain-text notation. Problem 5’s parenthetical note on superscript nesting (a{bc} = a{(bc)}) is about typography that plain text cannot show, so its distinction is visible only in the typeset original.
imos_2011	The six problems of the 2011 International Mathematical Olympiad, each marked out of 7. Contestants sat problems 1-3 on the first day and 4-6 on the second, with four and a half hours for each paper. The problem statements are taken from the official English paper. NOTE ON NOTATION: the text is extracted from a PDF, so mathematical layout is flattened – superscripts and fractions are lost, so n squared reads as ‘n2’ and one-over-n as ‘1 n’. Use the official PDF if you need the exact statement.
imos_2012	The item text is the organisers’ official English version of the IMO 2012 paper. The IMO is sat in each contestant’s own language, so most of the 547 contestants read a translation of this wording rather than the English shown here. Typeset mathematics is rendered in plain-text notation.
imos_2013	The item text is the organisers’ official English version of the IMO 2013 paper. The IMO is sat in each contestant’s own language, so most of the 527 contestants read a translation of this wording rather than the English shown here. Typeset mathematics is rendered in plain-text notation.
imos_2014	The item text is the organisers’ official English version of the IMO 2014 paper. The IMO is sat in each contestant’s own language, so most of the 560 contestants read a translation of this wording rather than the English shown here. Typeset mathematics is rendered in plain-text notation. Problem 6’s printed ‘Note:’ paragraph on partial credit is appended to that item’s text.
imos_2015	The item text is the organisers’ official English version of the IMO 2015 paper. The IMO is sat in each contestant’s own language, so most of the 577 contestants read a translation of this wording rather than the English shown here. Typeset mathematics is rendered in plain-text notation.
imos_2016	The six problems of the 2016 International Mathematical Olympiad, each marked out of 7, three on each of two days. The statements were read from images of the official English paper because its PDF text layer is corrupted, which means the mathematical notation here is more faithful than in the other Olympiad years in this batch – superscripts and subscripts are preserved.
imos_2017	The item text is the organisers’ official English version of the IMO 2017 paper. The IMO is sat in each contestant’s own language, so most of the 615 contestants read a translation of this wording rather than the English shown here. Typeset mathematics is rendered in plain-text notation.
imps2025_hf	Item text is blank by design: Hearts and Flowers stimuli are pictures (a red heart or a red flower), and the item code names the stimulus and the side it appeared on. The instructions and rule prompts are the canonical description of the standard task (Wright & Diamond 2014), not this study’s own script, which was never published. This study administered the task to groups of children on tablets with side-mounted response buttons and a 750 ms (spring: 1,250 ms) response deadline, whereas the canonical description is of an individually administered touchscreen version.
iowa_gambling_task	A card-choice task. On each trial the participant picked a card from one of four decks; two decks lose money in the long run and two gain. The item is one trial, and the response records WHICH DECK was chosen (1 = A, 2 = B, 3 = C, 4 = D) rather than whether the choice was right – there is no right answer to a single draw. Three versions of the task are pooled here, of 95, 100 and 150 trials, and the description of each item says which. TWO THINGS TO KNOW: the item numbers are not the trial numbers, because the processing script numbered items sequentially across the three pooled studies – the trial number is given in the item description instead. And each study is missing its final trial, lost when the source files were read.
ipip_openpsychometrics_ad	Ten items from the International Personality Item Pool’s Adventurousness scale, answered from 1 (strongly disagree) to 5 (strongly agree). Each item completes the stem ‘I …’. They were given as part of an online personality test alongside three other IPIP scales. A sixth response option meaning ‘not answered’ was recorded as 0 in the raw data and removed before this table was built.
ipip_openpsychometrics_sc	Ten items from the International Personality Item Pool’s Social Confidence scale, answered from 1 (strongly disagree) to 5 (strongly agree). Each item completes the stem ‘I …’. They were given as part of an online personality test alongside three other IPIP scales. A sixth response option meaning ‘not answered’ was recorded as 0 in the raw data and removed before this table was built.
isi_insomnia_wang2025	The item text here is the administered Chinese, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. This is an adapted Chinese Insomnia Severity Index: its interference item names study efficiency rather than the ISI original’s work and chores, so the administered wording is translated rather than replaced with the published English.
islam_2022_bsmas	The BSMAS was administered in Bengali, but the study published neither the Bengali wording nor an item-to-code key, so the item text is the standard English BSMAS assigned to BSMAS1-BSMAS6 by the instrument’s standard numbering; item 6’s “job/studies” reflects the original scale rather than what these school-going adolescents read.
islam_2022_gad7	The GAD-7 was administered in Bangla, but no Bangla wording is published, so the item and option text is the publisher’s official English. Item codes follow the official GAD-7 numbering (GAD4 = trouble relaxing, GAD5 = restlessness), although the paper’s Methods lists those two in the opposite order, and the response data cannot settle which is right.
islam_2022_phq9	The PHQ-9 was administered in Bangla, but no Bangla wording is published, so the item and option text is the official English form assigned to PHQ1-PHQ9 by its standard numbering. Separately, item PHQ9 (thoughts of self-harm) has no ‘Several days’ responses and 89 of 428 (20.8%) ‘Nearly every day’, a pattern present in the source deposit that looks like a coding problem.
iwasa_2016_padua_inventory	This study administered a Japanese translation of the Padua Inventory. The item and option wording shown here is the English original, not the Japanese text respondents actually saw: the source paper and its supplements do not reproduce the administered Japanese wording.
jablonska_2020_instagram_addiction	The item codes are an English rendering of the Polish taken from the data file’s headers and can contradict the wording: code 5 reads ‘Viewing Instagram is a good way to get bored’ where the administered Polish and the study’s English questionnaire both say it relieves boredom. Read the codes as identifiers only. The Polish anchors are aligned to resp 1-7 by the order the appendix prints them in.
jeilani_2024_self_efficacy	Item SEF4 has no published wording anywhere in this study’s deposit: it is the one self-efficacy item the SPSS file omits. Because the other four items are verbatim General Self-Efficacy Scale items 1, 6, 8 and 9 at their canonical numbers, SEF4 ships the canonical GSE item 4 wording (‘I am confident that I could deal efficiently with unexpected events.’) on the inference that the item code’s number is the GSE item number; the other four items’ text is transcribed from the deposit itself.
jeilani_2024_social_outcomes	Despite its name, this table is not a social-outcomes measure: its four items are the Significant Other subscale of the Multidimensional Scale of Perceived Social Support (Zimet et al., 1988), administered on a 5-point agreement scale rather than the original 7-point.
jeon_2019_cbi	On BT_W_4 (‘Do you have enough energy left to spend leisure time with friends or family?’) the response scale runs the opposite way to the other 18 items: the deposited data already store this item reverse-scored, so resp=4 corresponds to the questionnaire’s 0% anchor and resp=0 to its 100% anchor, and the shipped option_text reflects that. Averaging the items as stored reproduces the paper’s published subscale means exactly (38.59 / 33.94 / 34.88); reversing BT_W_4 first would give 36.12 for the work-related subscale. The English _translated columns are IRW’s own translation of the Korean questionnaire, not the authors’ or the CBI publisher’s.
jeon_2019_cesd10	The CESD-10 was administered to these Korean homecare workers in Korean, and no Korean wording is published in the article or either of its supporting files, so the text shown here is English. It is also not full item sentences: the study describes its ten items only as short phrases (‘loss of interest’, ‘trouble concentrating’, ‘feel pretty good’), and those phrases are what ship, because this Korean short form’s item 1 is ‘loss of interest’ rather than the canonical CES-D-10 item 1 and substituting the standard English sentences would have misrepresented the instrument. Items 5 and 8 are the positively worded ones and are stored unreversed in the response data, so the anchors shown apply as printed to every item
jian-wen-low-sepsis-and-trauma-resuscitation-2024	Items are rater-observed actions in two simulated resuscitation scenarios (scenario1 = sepsis/severe pneumonia, scenario2 = major trauma), each scored by two raters (rater column) as 0 / 0.5 / 1.
jiang_2024_growthm	Item text is the Chinese wording respondents read, taken from the study’s S1 Data SPSS variable labels; the English in the _translated columns was produced by IRW, not by the authors. Note also that the paper describes this scale as four items with a fixed-mindset example (“My intelligence is something that I can’t change very much”), while the table holds five items, all positively worded growth statements, and that example item is not among them
jiang_2024_instituinteg	Item text is the Chinese wording respondents read, taken from the study’s S1 Data SPSS variable labels; the English in the _translated columns was produced by IRW, not by the authors. Note also that the paper’s Measures section says the institutional integrity scale ‘has three questions’, while the table and the study’s own SPSS file both hold five
jiang_2024_ptsacc	Item text is the Chinese wording respondents read, taken from the study’s S1 Data SPSS variable labels; the English in the _translated columns was produced by IRW, not by the authors
jiang_2024_ptsexp	Item text is the Chinese wording respondents read, taken from the study’s S1 Data SPSS variable labels; the English in the _translated columns was produced by IRW, not by the authors.
jiang_2024_ptsinv	Item text is the Chinese wording respondents read, taken from the study’s S1 Data SPSS variable labels; the English in the _translated columns was produced by IRW, except for item PTSInv1, whose English is the authors’ own example from the paper.
jiang_2024_ptspr	Item text is the Chinese wording respondents read, taken from the study’s S1 Data SPSS variable labels; the English in the _translated columns was produced by IRW, not by the authors.
jimenezherrera_2022_moral_sensitivity	The English in the _translated columns is the original English wording of Lutzen’s Moral Sensitivity Questionnaire, printed by the paper beside its Spanish adaptation rather than translated from it, so it diverges from the administered Spanish in items 3, 7 and 9 (the authors replaced “to feel” with “to know” in item 3 and “balance” with “possibility” in item 7); the Spanish item_text is what respondents actually read, and it was transcribed from an image of the paper’s Table 1
jordan_2020_burnout	For Q12_8, Q12_9 and Q12_10 the study used two anchor sets across administrations (Never/almost never..Always on most dates, Never..Very Often on 7/31-8/3/2018). option_text is the majority wording, but 54/41/87 of the resp = 2 rows on those items come from the other anchor set, whose other levels were dropped in processing.
jovanovic_2026_conscientiousness_yips	This study administered a Serbian translation of the YIPS; the item text shown is the originator’s English wording in the originator’s item order, and that yips1-yips10 follow that order is assumed from the column numbering rather than confirmed against the data.
jung_2018_media_use	The survey was run in Korean, Chinese and Japanese (the questionnaires are in the S2-S4 Files, question B3); the shipped wording is the study’s own English from the paper’s Methods, because one table cannot carry three administered wordings. The three response labels are the paper’s collapsed coding: respondents saw four frequency options (not at all, 1-2, 3-4, 5 or more times), so resp=3 ‘more than three times’ merges the top two administered categories.
jutte_2024_loneliness	The three revised UCLA Loneliness items were administered in German to a Germany-wide panel, but neither the study’s SPSS deposit (no variable or value labels) nor the paper’s supplements publish the German wording, so the item text here is the paper’s own English rendering rather than what respondents read. The paper describes the scale as 1 = never to 7 = very frequently while the table stores 0-6, so the anchors are shipped as 0 = never to 6 = very frequently; the five intermediate points are unlabelled in the source.
jutte_2024_personality	The Big Five single items were administered in German, but the German wording is not published in the deposit or the paper’s supplements, so the item text here is the English rendering the paper prints in its Table 1 rather than what respondents read. The paper states a 1-7 scale while the table stores 0-6; only the endpoints are labelled.
K-PCQ_Huh_2022	The item text is the administered Korean from the study’s own deposited questionnaire (Supplement 1), with the study’s English alongside it. Readers going back to the source paper should note that its Table 1 lists the same twelve items in a different order from the questionnaire the data follow: the paper’s Table 1 item 1 is this table’s Crav5, its item 3 is Crav11, and so on. The ordering shipped here is the questionnaire’s, and it is what the paper’s own Table 2 correlations and its ‘item 11’ misfit statement refer to
kanwal_2024_csr	The wording of item CSRN3 is cut off in the source’s own item table (it ends ‘… eco- design,’), so its item_text is incomplete.
karajko2025_ai_benefit	The item text, response options and block prompt here are the administered Bosnian/Croatian, transcribed from the study’s own questionnaire. The English in the _translated columns is a machine translation produced by IRW – it is not the study’s wording, and it has not been reviewed or back-translated by a Bosnian/Croatian speaker. No English version of this block is published in the source record (ESM_4.pdf is entirely Croatian, and the spreadsheet’s parenthetical English glosses cover only an unrelated block). Treat the translation as a reading aid rather than a validated English instrument.
karajko2025_ai_governance	The item text, response options and block prompt here are the administered Bosnian/Croatian, transcribed from the study’s own questionnaire. The English in the _translated columns is a machine translation produced by IRW – it is not the study’s wording, and it has not been reviewed or back-translated by a Bosnian/Croatian speaker. No English version of this block is published in the source record (ESM_4.pdf is entirely Croatian, and the spreadsheet’s parenthetical English glosses cover only an unrelated block). Treat the translation as a reading aid rather than a validated English instrument.
karajko2025_ai_risk	The item text, response options and block prompt here are the administered Bosnian/Croatian, transcribed from the study’s own questionnaire. The English in the _translated columns is a machine translation produced by IRW – it is not the study’s wording, and it has not been reviewed or back-translated by a Bosnian/Croatian speaker. No English version of this block is published in the source record (ESM_4.pdf is entirely Croatian, and the spreadsheet’s parenthetical English glosses cover only an unrelated block). Treat the translation as a reading aid rather than a validated English instrument.
karajko2025_ai_trust	The item text, response options and block prompt here are the administered Bosnian/Croatian, transcribed from the study’s own questionnaire. The English in the _translated columns is a machine translation produced by IRW – it is not the study’s wording, and it has not been reviewed or back-translated by a Bosnian/Croatian speaker. No English version of this block is published in the source record (ESM_4.pdf is entirely Croatian, and the spreadsheet’s parenthetical English glosses cover only an unrelated block). Treat the translation as a reading aid rather than a validated English instrument.
karpudewan_2022_stp_cca	The paper describes a five-point response scale, 1 (strongly disagree) to 5 (strongly agree), but no response of 5 occurs anywhere in the authors’ data (11,513 responses here and 9,900 in the companion EFA sample are all 1-4), so the form as administered appears to have been four- point and the response labels are left blank rather than attached to the wrong levels. The instructions field is the paper’s own sentence describing the task, not the questionnaire’s instruction text, which is not published.
kermen_2022_self_efficacy	The 11 reverse-scored items of this scale are stored already reversed, so the response anchors shipped for them run in the opposite direction to the other six: for b2, b4, b5, b6, b7, b10, b11, b12, b14, b16 and b17 a stored resp of 1 is the administered answer ‘cok iyi’ (very well) and 5 is ‘hic’ (not at all). The English item text was translated by IRW from the published Turkish form; only the two scale endpoints were labelled in the source, so resp 2-4 carry no option text
kern_2021_connectedness	Despite the table name, these nine items are the study’s ‘Course design’ evaluation statements (Gan et al. 2022), not a connectedness scale; only the scale endpoints (1 = Strongly disagree, 4 = Strongly agree) were published.
kern_2021_engagement	Despite the table name, these nine items are the study’s ‘Educator style’ evaluation statements (Gan et al. 2022), not an engagement scale; only the scale endpoints (1 = Strongly disagree, 4 = Strongly agree) were published.
kern_2021_learning	Despite the table name, these twelve items are the study’s ‘Life skills’ learning-outcome statements (Gan et al. 2022); only the scale endpoints (1 = Strongly disagree, 4 = Strongly agree) were published.
khattak_2026_confidence	The study’s questionnaire prints a single ‘Yes / No’ header over all eight rows of its Section Four, including these four confidence items, while the paper’s Methods and Table 3 describe them on a ‘Very confident / Not confident’ scale; the option text follows the paper.
khoa_2023_assurance	The item codes (ASS) stand for Academic Staff Satisfaction in the source questionnaire, not ‘assurance’ as the table name suggests.
kim_2023_gad7	The GAD-7 was administered in Korean, but no Korean wording is published in the paper or its SPSS deposit, so the item and option text here is the instrument publisher’s official English (phqscreeners.com) rather than the sentences respondents read. The shipped instruction is the GAD-7’s own ‘Over the last 2 weeks’, while the paper’s Methods describes its GAD-7 as asking about anxiety over the past two months. The deposit’s columns are unlabelled, so codes are tied to wording by the form’s printed numbering; data checks pin gad5 (restlessness) and gad6 (irritability) but do not separate the remaining items.
kim_2023_phq9	The PHQ-9 was administered in Korean (the Lee et al. 2014 Korean adaptation), but no Korean wording is published in the paper or its SPSS deposit, so the item and option text here is the instrument publisher’s official English (phqscreeners.com) rather than the sentences respondents read. The shipped instruction is the PHQ-9’s own ‘Over the last 2 weeks’, while the paper’s Methods describes its PHQ-9 as asking about depression experiences over the past two months. The deposit’s columns are unlabelled, so codes are tied to wording by the form’s printed numbering; data checks pin PHQ9_4 (fatigue), PHQ9_9 (self-harm) and the somatic/cognitive split but do not separate the remaining items.
kim_2025_isi	The Insomnia Severity Index was administered in Korean, but no Korean wording exists in the deposit or the paper, so the text is English from two different sources: the seven item stems are the published English ISI wording from Lenderking et al. (2024), not a translation of what these respondents read, while the option anchors are the study’s own English labels from its S1 File. Response codes run 1-5, not the ISI’s published 0-4, so a total summed from the raw resp values is inflated by 7 relative to the standard 0-28 ISI total. Item 4’s anchors as the study labelled them run ‘completely unaware’ to ‘fully aware’ rather than the standard form’s ‘not at all noticeable’ to ‘very much noticeable’. No source ties a code to a stem; the mapping is reconstructed from the S1 File’s per-column value labels and same-file cross-checks.
kim_2025_psas	The Korean Pre-Sleep Arousal Scale was administered in Korean, but no Korean wording appears in the paper or its single supplementary file, so the item text here is the English wording the authors print in their Table 3 rather than the sentences respondents read. The response labels are the deposit’s own legend (1 = not at all … 5 = very often), which differs from the paper’s Methods sentence describing the scale as running from 1 (not at all) to 5 (extremely).
kim2020_ams	The study published no item wording, so the shipped Korean text is the official Korean Aging Males’ Symptoms form (Heinemann et al. 2003), not the study’s own materials, and the English in the _translated columns is the AMS’s official English version rather than a translation of the Korean. The form is aligned to the data’s Q01-Q17 by the AMS’s canonical item numbering; that alignment reproduces the scale’s three-factor structure but does not distinguish items within a subscale.
kiraly_2024_perinatal_mh_freq	Four items were asked on both the obstetrician and the pediatrician/NP survey forms with slightly different wording, and each code carries only one: the variant the deposit’s column name reproduces. Respondents to the other form read the other wording.
kiraly_2024_perinatal_mh_symptoms	The 19 items are the checkboxes of a single survey question, so option_text ‘yes’/‘no’ records whether the symptom was selected rather than an anchor respondents read, and the instructions text is the stem as worded on the study’s obstetrician survey form – the parallel pediatrician/nurse-practitioner form, which 85 of the 101 respondents completed, is not publicly retrievable and may have worded that stem slightly differently
kitayama_2022_hweat	The Japanese item wording is the study’s own; the English in the item_text_translated column was produced by IRW rather than by the study’s authors, and the administered Japanese response anchors were never published, so option_text is blank and only the paper’s English endpoint descriptions (1 = strongly disagree, 5 = strongly agree) are carried.
kohlmann_2016_phq9	The study’s German PHQ-9 form was not published, so the item text is the standard German PHQ-D wording, with the official English PHQ-9 alongside; that the study used this wording is inferred, not documented. The item order was checked against the paper’s per-symptom frequencies.
kokoszka_2022_hamd	Item text is a canonical public-domain English reproduction of Hamilton’s HDRS-17, not the wording used here: the scale was rated in Polish by psychiatrists following the SIGH-D interview guide, and the deposit (a bare CSV of integers) publishes no Polish wording. Codes are matched to the instrument’s own numbering. ham16a and ham16b are the HDRS’s two alternate ratings of the single item 16 (weight loss by patient report vs by weekly measurement), not two items; both are scored for every patient, but only ham16b enters the paper’s HAM-D total.
komura_2026_gqs_animacy	Respondents answered in Japanese; the item text is the authors’ own English translation from their S2 File, because no Japanese wording is published anywhere in the deposit. The block is a 5-item variant of Godspeed Animacy, not the canonical 6-item subscale – ‘Inactive <-> Active’ and ‘Sleepy <-> Awake’ are not canonical Animacy pairs – so the wording is this study’s own, and no published Godspeed wording was substituted.
komura_2026_gqs_anthropomorphism	Respondents answered in Japanese; the item text is the authors’ own English translation from their S2 File, because no Japanese wording is published anywhere in the deposit. The anchors are also an adaptation rather than the canonical Godspeed Anthropomorphism set: items 4 and 5 (Still-Lively, Mechanical-Organic) are Godspeed animacy anchors, and the canonical Fake-Natural and Moving rigidly-Moving elegantly pairs were not used. The study’s wording is shipped as given, not normalised to the published instrument.
komura_2026_gqs_likeability	Respondents answered in Japanese; the item text is the authors’ own English translation from their S2 File, because no Japanese wording is published anywhere in the deposit. That translation departs from the published Godspeed Likeability subscale, and is shipped as given: items 1 and 3 carry the identical text “Unpleasant <-> Pleasant” (most likely two distinct Japanese pairs rendered into one English phrase, so the two cannot be told apart), and item 5 is “Scary <-> Not scary”, which belongs to no Godspeed subscale. Only item 5 is pinned to its code by the data; the order of items 1-4 rests on the S2 File’s numbering.
komura_2026_gqs_perceived_intelligence	Respondents answered in Japanese; the item text is the authors’ own English translation from their S2 File, because no Japanese wording is published anywhere in the deposit. The block departs from the published Godspeed Perceived Intelligence subscale: the canonical fifth pair “Foolish - Sensible” is replaced by “Unresponsive <-> Responsive”, a Godspeed animacy anchor the study also administered as Animacy item 4. Items 1, 3 and 5 (Competent, Knowledgeable, Intelligent) are not separated by the data, so their order rests on the S2 File’s numbering.
komura_2026_gqs_perceived_safety	Respondents answered in Japanese; the item text is the authors’ own English translation from their S2 File, because no Japanese wording is published anywhere in the deposit. It departs slightly from the published Godspeed Perceived Safety subscale, shipped as given: “Calm” where the canonical set has “Relaxed”, “Peaceful” where it has “Quiescent”, and item 2 printed as “Calm - Agitated”, so “Calm” is the high pole of item 1 and the low pole of item 2. Responses are stored raw: items 2 and 3 are reverse-keyed and must be reversed before scoring.
komura_2026_mdmt_capable	Respondents answered in Japanese; the item text is the authors’ own English translation from their S2 File, because no Japanese wording is published anywhere in the deposit. The paper cites Lee & See’s trust-in-automation work, but the items are verbatim the Capable subscale of Ullman & Malle’s Multi-Dimensional Measure of Trust. Only item 4 (Meticulous) is pinned to its code by the data; items 1-3 rest on the S2 File’s numbering.
komura_2026_mdmt_ethical	Respondents answered in Japanese; the item text is the authors’ own English translation from their S2 File, because no Japanese wording is published anywhere in the deposit. The wording is the study’s rather than the canonical MDMT Ethical subscale (‘Has integrity’ is third and ‘Honest/upright’ fourth). The four near-synonymous adjectives share one 0-7 scale and the paper publishes no per-item statistics, so their assignment to mdmt_ethical_1..4 follows the S2 File’s numbering and is not confirmed against the data.
komura_2026_mdmt_reliable	Respondents answered in Japanese; the item text is the authors’ own English translation from their S2 File, because no Japanese wording is published anywhere in the deposit. It departs from the published MDMT Reliable subscale at one position, shipped as given: item 3 is “Dependable” where the canonical subscale reads “someone you can count on”. The data do not separate item 1 from item 3 or item 2 from item 4, so order within the block rests on the S2 File’s numbering.
komura_2026_mdmt_sincere	Respondents answered in Japanese; the item text is the authors’ own English translation from their S2 File, because no Japanese wording is published anywhere in the deposit. The four sincerity words are near-synonyms and the paper reports no per-item statistics, so their order within the block rests on the S2 File’s numbering alone.
konerding_2019_patientsatisfaction	Item wording is the ENGLISH, diabetes-survey version printed in Table 1 of the source paper. The questionnaire was administered in six languages (English, Finnish, German, Greek, Dutch, Spanish; all six wordings are in that table), and stroke-survey respondents saw the same items rephrased in the past tense and referring to the treating hospital rather than to diabetes-related services, so respondents outside the English diabetes sample read a translated or re-referred variant of the text shipped here. Scale points 2-6 are unlabelled in the source; only the 1 and 7 anchors are shipped.
KoreanNursing_Park_2017	Item text for this table is the study authors’ English rendering in Supplement 7 of the paper; the mock examination itself was administered in Korean and no Korean wording is published in the article, its supplements or the Harvard Dataverse deposit. Three things about this table are easy to misread: resp runs 1 = INCORRECT and 0 = correct (confirmed item by item against Supplement 7’s published correct-answer counts, all 50 exact); the item codes X0, X1, X0.1 … encode nothing about the items but are R’s mangled column names from a headerless raw file whose first examinee was consumed as the header, so the table holds 740 of the study’s 741 respondents; and the five answer options appear inside item_text rather than in option_text, which carries the 0/1 accuracy labels instead. The study published no answer key, so correct_response was derived by IRW from Supplement 7’s own upper/lower-27% response distributions (unique for 49 of the 50 items; item 5 from its 84% correct-answer rate).
kotsou_2016_panas	Respondents read a French version that the source does not publish or identify, so the item and response-option wording is the standard English PANAS (Watson, Clark & Tellegen, 1988), assigned by item number. The data confirm each item’s positive/negative affect grouping but not which adjective is which within a group. The instructions are not published and are left blank.
kraft_todd_2017_competence	The live resp values 1-5 are the survey’s 1-based codes for a scale the questionnaire displayed as 0 (not at all) to 4 (very much), so option_text shows both; the paper’s reported competence means (3.64 empathic vs 3.21 unempathic) are on the stored 1-5 coding, and the form’s sixth option ‘Does not apply’ is stored as missing rather than as a resp level
kraft_todd_2017_panas	The item-to-text alignment is inferred from the order items appear in the source, not from an explicit code-to-text mapping
kraft_todd_2017_warmth	The questionnaire printed this scale as 0 (not at all) to 4 (very much), but the data store it as 1-5, so resp 1 is ‘not at all’ and resp 5 ‘very much’; the ‘Does not apply’ option is stored as missing. The adjectives are assigned to WARM1-WARM4 by the questionnaire’s listing order, and ‘Sincere’ (WARM3) and ‘Good natured’ (WARM4) could not be told apart in the data.
kuczyk_2024_facemask_fba	Item text is the authors’ own English rendering from the variable labels of their SPSS file; the questionnaire was administered in German and no German wording is published in the deposit or supplements. The response anchors are the .sav’s value labels, which differ in wording from the paper’s Methods gloss (1 = strongly disagree … 5 = strongly agree); both are the authors’ English for the same German scale.
kuczyk_2024_facemask_fbe	Item text is the authors’ own English rendering from the variable labels of their SPSS file; the questionnaire was administered in German and no German wording is published in the deposit or supplements (the authors say the questionnaires are available on request). The response anchors are the .sav’s value labels (1 = definitely disagree … 5 = definitely agree), which the paper’s Methods renders instead as 1 = strongly disagree … 5 = strongly agree.
kuehner_2017_mw_rumination	The two probes were administered in German to students in Mannheim, but no German wording is published in the paper or any of its supplementary files, so the item text is the authors’ own English from the paper’s Methods rather than the sentences respondents read.
kumlander_2018_scs	English for the three middle response anchors (resp 2, 3 and 4) was rendered by IRW from the Finnish; the study published English only for the two endpoints, and the Finnish in option_text is what respondents actually read.
kushnir2017_anrt	The item code is the source file’s full column header, so the join key contains the item wording itself; item_text is that same wording with the “(ANRT_n)” prefix removed. Respondents could also answer “Don’t know”, which the source drops rather than codes, so between 8% and 16% of responses per item are absent from the table and are not missing at random – the five levels here come from six that were offered. The stem that says what “these products” refers to is not published in the source record; per the IRW dictionary the referent is nicotine replacement therapy, but that is not transcribed text.
laksmita_2020_mspss	The study administered an Indonesian translation of the MSPSS, but neither the paper nor its S1 Dataset publishes the Indonesian wording, so the canonical English MSPSS is shipped in its place, including the canonical form’s instructions and its anchors for points 2-6 (the paper states only the 1 and 7 endpoints). Where the paper’s own English differs from the canonical form (item 2, ‘share my joys and sorrows’), the canonical wording is shipped.
latifi_2026_insect_fear	The wording is the study’s own English questionnaire (S1 Table); the children answered a Persian version the article does not publish. The response scale shipped for Q1-Q23 runs 1 = Strongly agree to 5 = Strongly disagree, the OPPOSITE of the direction in both the SPSS deposit’s value labels and the Methods section; the data force it – only this direction reproduces the paper’s published age correlations. Q24-Q32 run 1 = Never to 5 = Always.
lec5_xiong_2025	The item wording is the English LEC-5 Standard form from the US National Center for PTSD, not the Chinese version these four samples completed, which the study did not publish. Item order is assumed to follow the standard LEC-5 numbering. resp is the study’s own 0/1 exposure indicator (1 = exposed to the event), not one of the LEC-5’s six response categories, and the rule used to collapse those categories is not documented.
lee_2020_alcohol_use	This is a past-year, past-tense AUDIT-C variant from the study’s own data file, not the WHO AUDIT’s present-tense wording (AUDIT: Babor et al., World Health Organization, 2001). For Drinking_habit_2 the response values are the questionnaire’s six categories in order (0 = 0 drinks .. 5 = 10 or more drinks), not AUDIT-C points; standard scoring gives both ‘0 drinks’ and ‘1-2 drinks’ 0 points. Drinking_habit_1’s highest category is ‘4-5 times a week’ rather than the standard ‘4 or more times a week’.
lee_2020_vr_usability	Administered in Korean to Korean nursing students, but the Korean wording is not published, so IRW ships the English the paper prints in its Table 2. One Usability7 response (participant M12) is 10 in the deposited data, while the paper’s statistics for that item (mean 2.60, SD 2.61, range 0-8) are reproduced only if that response is 0. No response anchors were published.
lee_2024_panas	The PANAS was administered in Korean (the Park & Lee, 2016 translation), but neither the PeerJ paper nor its supplements reproduce any Korean wording – the deposit workbook carries bare column headers PANAS1..PANAS20 and no variable labels – so the item text shown is the English PANAS adjectives and is not what respondents read. Note that PANAS1..PANAS20 do NOT follow the canonical Watson, Clark & Tellegen (1988) item order: the Korean version runs interested, irritable, distressed, alert, excited, ashamed, upset, inspired, strong, nervous, guilty, determined, scared, attentive, hostile, jittery, enthusiastic, active, proud, afraid, so positive-affect items sit at positions 1, 4, 5, 8, 9, 12, 14, 17, 18 and 19. The live data confirms this (mean within-valence correlation 0.40 against 0.00 between valences, versus 0.20 and 0.17 under the canonical order), but the scared/afraid pair (PANAS13/PANAS20) cannot be told apart from the data alone. Only the two endpoints of the 1-5 scale are labelled in the source (‘not at all’, ‘very true’); the middle points are left blank rather than invented
lee_2025_nursing_exam	The Korean item and instruction text is transcribed verbatim from the study’s own deposited examination paper; the English in the _translated columns was produced by IRW, not by the study authors. Each multiple-choice item’s five alternatives are carried inside item_text, because the response data records a 0/1 correct-incorrect score rather than the alternative a student chose.
lee2024_relative_clause	The source test has 25 items; only items 1-18 are in this table. Items 19-25 were a separate grammaticality-judgement task with a different response format and were not carried over. The items are in English by design – this is a test of English relative clauses given to Korean high-schoolers – while the printed instructions carried Korean glosses, transcribed as printed. Items 1-3 are a picture-circling task and items 4-18 read-and-answer, each with its own administered direction, and item_text gives the stimulus sentence together with its comprehension question, as they were administered. Source typos are kept verbatim.
leon_guereno_2020_breq	The BREQ-3’s 0-4 response scale was administered here as 1-5, and the two descriptions of its anchors disagree: the study’s SPSS deposit labels the five points as an agreement scale while the paper states the instrument’s usual ‘not true for me’ / ‘sometimes true for me’ / ‘very true for me’ anchoring; the instrument’s anchors are what we ship, and the unlabelled scale points are left blank rather than filled in
leon_rubio_2025_cdpe_antecedents	The English translations (*_translated) were written by IRW from the Spanish CDPE questionnaire respondents read; the Spanish item text is the questionnaire as deposited.
leon_rubio_2025_cdpe_burnout	As for leon_rubio_2025_cdpe_antecedents: the English translations were written by IRW from the deposited Spanish questionnaire.
leon_rubio_2025_cdpe_consequences	As for leon_rubio_2025_cdpe_antecedents: the English translations were written by IRW from the deposited Spanish questionnaire.
leon_rubio_2025_cdpe_coping	As for leon_rubio_2025_cdpe_antecedents: the English translations were written by IRW from the deposited Spanish questionnaire.
leon_rubio_2025_cdpe_hardiness	As for leon_rubio_2025_cdpe_antecedents: the English translations were written by IRW from the deposited Spanish questionnaire.
lestari_2026_cognitive_flexibility	Response options are listed without their 0-3 scores because the authors did not publish which option earns which score, so option text cannot be joined to resp; item CF7 also shows a fifth, unedited placeholder option (‘Option 5’) exactly as the administered form did, and the English is IRW’s own translation.
lestari_2026_inhibitory_control	Response options are listed without their 0-3 scores because the authors did not publish which option earns which score, so the option text cannot be joined to resp; the English text is IRW’s own translation of the Indonesian form.
lestari_2026_working_memory	Response options are listed without their 0-3 scores because the authors did not publish which option earns which score, so the option text cannot be joined to resp; the English text is IRW’s own translation of the Indonesian form.
li_2021_cultural_intelligence	The questionnaire was administered to employees of Chinese enterprises, but the deposit publishes only an English version (S1 Appendix contains no Chinese text), so the item text is the authors’ own English rather than what respondents read. The administration language is inferred from the sample and survey platform; the paper does not state it. The data confirm which facet each item belongs to but not the order within a facet, which rests on the S1 Appendix numbering.
li_2021_knowledge_sharing	The survey was fielded to employees of Chinese enterprises and was in all likelihood administered in Chinese (the paper does not state the language), but the only questionnaire published is in English, so the item text is the authors’ own English rendering rather than the wording respondents read. The data do not separate KS2 from KS4, so their order rests on the S1 Appendix numbering.
li_2021_org_cultural_diversity	The survey was fielded to employees of 31 Chinese enterprises and was in all likelihood administered in Chinese, but the only questionnaire in the study’s supplements is in English, so the item text shown is the authors’ own English rendering rather than the wording respondents read; the questionnaire numbers its two sub-blocks separately (1-4 and 1-5), so the assignment of those items to OCD1-OCD9 follows listing order, which the data support only in part; the seven scale points are unlabelled boxes, so no response-option text is available.
li_2021_sustainable_innov_behav	The survey was fielded to employees of 31 Chinese enterprises and was in all likelihood administered in Chinese (the paper does not state the language), but the only questionnaire published is in English, so the item text is the authors’ own English rendering rather than the wording respondents read. The data do not order SIB4 against SIB5 or SIB1 against SIB3, so those rest on the S1 Appendix numbering; SIB4 and SIB5 are the two items the paper dropped from its analysis.
li_2024_bdyz	Despite the dictionary’s ‘body-image’ description, this is the ERQ expressive-suppression subscale. Item text is the ERQ’s original English; the study administered the scale in Chinese and no Chinese wording is published in the deposit or its supplements. The deposit numbers its four columns 1, 2, 6, 9 while the ERQ numbers its suppression items 2, 4, 6, 9, so bdyz_1 carries ERQ item 4 (“When I am feeling positive emotions, I am careful not to express them”) by elimination rather than a stated correspondence.
li_2025_corporate_performance	Item text for this table is the study authors’ English from Table 3 of the paper; the survey was administered in Chinese to employees of China Time-Honored Brand enterprises and no Chinese wording appears anywhere in the article or its supplement, so these are not the words respondents read. Two further caveats: the response-option labels shipped here (‘Very inconsistent’ … ‘Completely consistent’) are the .sav’s own value labels, which disagree with the article’s Measures section describing the same scale as 1 ‘strongly disagree’ to 7 ‘strongly agree’; and the instructions line is IRW’s second-person rendering of the paper’s Methods sentence rather than a published instruction, supplied because the ten item labels are bare performance indicators with no referent of their own
li_2025_market_environment	The five item sentences are the authors’ own English: the survey was administered in Chinese to staff at Chinese time-honored brand enterprises and no Chinese wording is published in the article or its single supplement. The response-option text is taken from the SPSS file’s own value labels (Inconsistent / General / Consistent / Very consistent / Completely consistent), which disagree with the paper’s prose description of the same scale as running from 1 “strongly disagree” to 7 “strongly agree”; two obvious typos in those labels were corrected. Scale points 1 and 2 were never used by any respondent, so no option text is shipped for them
li_2025_marketing_culture	The study surveyed managers and employees at Chinese time-honored-brand enterprises and the scale went through translation and back-translation, so it was administered in Chinese, but the only wording published anywhere in the article or its S1 File is English – the item text shown here is the authors’ own English rendering rather than what respondents read. The response-option text also disagrees with the article body: the paper describes the scale as running from 1 “strongly disagree” to 7 “strongly agree”, while the S1 File’s own value labels, which are what is shown here, read “Extremely inconsistent” through “Completely consistent”; both run in the same direction, and levels 1 and 2 carry no rows because no respondent used them
li_2025_marketing_exploitation	The five item sentences are the study authors’ own English, taken from the article’s measurement table: the survey was administered in Chinese to managers and employees of China Time-Honored Brand enterprises and no Chinese wording appears anywhere in the article or its single supplement, so these are not the words respondents read. The response-option text is the SPSS file’s own value labels (Inconsistent / General / Consistent / Very consistent / Completely consistent), which disagree with the article’s own description of the same scale as running from 1 “strongly disagree” to 7 “strongly agree”; both run in the same direction, two obvious typos in those labels were corrected, and scale points 1 and 2 carry no rows because no respondent used them
li_2025_marketing_exploration	The five item sentences are the study authors’ English: the survey was administered in Chinese to managers and employees of China Time-Honored Brand enterprises, and no Chinese wording is published in the article or in its only supplement, so these are not the words respondents read. The response-option text is taken from the SPSS file’s own value labels (“Very inconsistent” through “Completely consistent”), which disagree with the article’s Measures section describing the same scale as running from 1 “strongly disagree” to 7 “strongly agree”; both run in the same direction, and two obvious typos in those labels were corrected. Scale point 1 was never used by any respondent, so no option text is shipped for it.
li_2025_marketing_learning	The five item sentences are the authors’ own English: the survey was administered in Chinese to managers and employees at China Time-Honored Brand enterprises, with translation and back-translation, and no Chinese wording is published in the article or its single data supplement. The response-option text is taken from the SPSS file’s own value labels (Inconsistent / General / Consistent / Very consistent / Completely consistent), which disagree with the article’s Measures section describing the same scale as running from 1 “strongly disagree” to 7 “strongly agree”; both run low-to-high, and two obvious typos in those labels were corrected. Scale points 1 and 2 were never used by any respondent, so no option text is shipped for them
li_2025_marketing_operation	The five item sentences are the study authors’ own English, taken from the article’s measurement table: the survey was administered in Chinese to managers and employees of China Time-Honored Brand enterprises, with translation and back-translation, and no Chinese wording appears in the article or in its single data supplement, so these are not the words respondents read. The response-option text is the SPSS file’s own value labels (Inconsistent / General / Consistent / Very consistent / Completely consistent), which disagree with the article’s Measures section describing the same scale as running from 1 “strongly disagree” to 7 “strongly agree”; both run low-to-high, and two obvious typos in those labels were corrected. Scale points 1 and 2 were never used by any respondent, so no option text is shipped for them
li_2025_policy_environment	The five item sentences are the study authors’ own English, taken from the article’s measurement table: the survey was administered in Chinese to managers and employees of China Time-Honored Brand enterprises, with translation and back-translation, and no Chinese wording appears in the article or in its single data supplement, so these are not the words respondents read. The response-option text is the SPSS file’s own value labels (Inconsistent / General / Consistent / Very consistent / Completely consistent), which disagree with the article’s Measures section describing the same scale as running from 1 “strongly disagree” to 7 “strongly agree”; both run low-to-high, and two obvious typos in those labels were corrected. Scale points 1 and 2 were never used by any respondent, so no option text is shipped for them. Note also that the response table itself is uneven by design: PE2-PE5 each lost about 40 responses that the processing script dropped as corrupted (a multiplicative transformation had been applied to a subset of respondents, producing values such as 7.35 and 7.497 on a 1-7 scale), so PE1 has 352 rows and the other four have 312-313
li_2025_socmedia_enjoyment	This table was administered in Chinese and the English in the _translated columns comes from two different sources. item_text_translated is the study’s own English: the authors’ S1 File (Appendix A) prints an English measurement-item table for every item, and that is what ships. option_text_translated is only part theirs – the article prints English for the two ENDPOINTS of the five-point scale (‘strongly disagree’ / ‘strongly agree’) and for nothing in between, so the three intermediate anchors shown as ‘Disagree’, ‘Neutral’ and ‘Agree’ are IRW’s renderings of the administered Chinese 不赞同 / 中立 / 赞同 rather than the authors’ own words. The administered Chinese ships verbatim in item_text and option_text.
li_2025_socmedia_ewom	The item wording is the study’s own in both languages: administered Chinese from the deposit’s SPSS variable labels, and the authors’ own English from Appendix A. The response anchors are a different matter. The Chinese in option_text is the study’s, but the article prints only the two endpoints in English (“strongly disagree” and “strongly agree”) and the supporting file publishes no English anchors at all, so “Disagree”, “Neutral” and “Agree” in option_text_translated are IRW’s renderings of 不赞同 / 中立 / 赞同 rather than the authors’.
li_2025_socmedia_infoquality	This table was administered in Chinese and the English in the _translated columns comes from two different sources. item_text_translated is the study’s own English: the authors’ S1 File (Appendix A) prints an English measurement-item table for every item, and that is what ships. option_text_translated is only part theirs – the article prints English for the two ENDPOINTS of the five-point scale (‘strongly disagree’ / ‘strongly agree’) and for nothing in between, so the three intermediate anchors shown as ‘Disagree’, ‘Neutral’ and ‘Agree’ are IRW’s renderings of the administered Chinese 不赞同 / 中立 / 赞同 rather than the authors’ own words. The administered Chinese ships verbatim in item_text and option_text.
li_2025_socmedia_revisit	This table was administered in Chinese and the English in the _translated columns comes from two different sources. item_text_translated is the study’s own English: the authors’ S1 File (Appendix A) prints an English measurement-item table for every item, and that is what ships. option_text_translated is only part theirs – the article prints English for the two ENDPOINTS of the five-point scale (‘strongly disagree’ / ‘strongly agree’) and for nothing in between, so the three intermediate anchors shown as ‘Disagree’, ‘Neutral’ and ‘Agree’ are IRW’s renderings of the administered Chinese 不赞同 / 中立 / 赞同 rather than the authors’ own words. The administered Chinese ships verbatim in item_text and option_text.
li_2025_socmedia_satisfaction	This table was administered in Chinese and the English in the _translated columns comes from two different sources. item_text_translated is the study’s own English: the authors’ S1 File (Appendix A) prints an English measurement-item table for every item, and that is what ships. option_text_translated is only part theirs – the article prints English for the two ENDPOINTS of the five-point scale (‘strongly disagree’ / ‘strongly agree’) and for nothing in between, so the three intermediate anchors shown as ‘Disagree’, ‘Neutral’ and ‘Agree’ are IRW’s renderings of the administered Chinese 不赞同 / 中立 / 赞同 rather than the authors’ own words. The administered Chinese ships verbatim in item_text and option_text.
li_2025_socmedia_usefulness	This table was administered in Chinese and the English in the _translated columns comes from two different sources. item_text_translated is the study’s own English: the authors’ S1 File (Appendix A) prints an English measurement-item table for every item, and that is what ships. option_text_translated is only part theirs – the article prints English for the two ENDPOINTS of the five-point scale (‘strongly disagree’ / ‘strongly agree’) and for nothing in between, so the three intermediate anchors shown as ‘Disagree’, ‘Neutral’ and ‘Agree’ are IRW’s renderings of the administered Chinese 不赞同 / 中立 / 赞同 rather than the authors’ own words. The administered Chinese ships verbatim in item_text and option_text.
li_2026_imi_teq	The survey was administered in Chinese – the study is a forward-backward translation of the IMI Task Evaluation Questionnaire validated on 362 Chinese students – but the Figshare deposit’s only file is titled “Dataset English.xlsx” and contains no Chinese wording, so the sentences shown here are the authors’ English rather than what respondents read. Two further caveats: the eight reverse-worded items (item_07, item_15, item_16, item_17, item_18, item_19, item_26, item_30) are stored already reverse-scored, each correlating positively (+0.27 to +0.63) with the forward items of its own subscale, so a high response means low agreement with the sentence as printed; and no response-option wording is shipped, because neither the deposit nor any locatable publication states the scale anchors. Note also that the table holds 34 items, not the 22 of the IMI-TEQ: items 1-22 are the IMI-TEQ re-ordered into subscale blocks, items 23-32 a ten-item learning-fulfillment scale and items 33-34 two system-usability items
li_2026_psmus	The source publishes its item wording only in English, although the questionnaire was administered in Chinese.
li_2026_sas_sv	The questionnaire was administered in Chinese, but neither the paper nor its data deposit publishes the Chinese wording, so the item text is the authors’ English from the paper’s Table 2 (the standard Kwon et al. 2013 SAS-SV items). The study states only the two endpoint anchors (1 = strongly disagree, 6 = strongly agree), so response options 2-5 carry no label.
liang_2026_extrinsic_motivation	Item wording for this table comes from the paper’s Table 1, which numbers the ten items 1-10 and groups items 6-10 as extrinsic motivation but never prints the EM1-EM5 codes the data uses, so the assignment EM1=item 6 … EM5=item 10 follows the questionnaire’s presentation order. The live correlations support it as far as they can: the three items the paper adapted from one source scale intercorrelate 0.77-0.83 while the two single-source items sit lower (item 9 at most 0.68, item 10 at most 0.57, item-rest 0.59), so those three are distinguished from the other two, but a permutation among the first three would not be detectable. The response coding is not inferred – the deposit ships each item as both a label and an integer, and the two agree cell for cell (2=disagree, 3=ordinary, 4=agree, 5=strongly agree; 1=strongly disagree was never used). The questionnaire was administered in Chinese; only the authors’ English is published, so that is what ships.
liang_2026_intrinsic_motivation	The questionnaire was administered in Chinese and only the authors’ English is published, so English is what ships. The paper never prints the IM1-IM5 codes; the assignment follows Table 1’s order and is confirmed item by item against the per-item means and SDs in Table 8. The paper prints item 2 two ways (Table 1: “I would use the app to learn more about sports if time permits”; Table 8: “If time permits, I would use …”); the Table 1 form ships.
liem_2024_attitude_env	Two things to know before using this table. First, the seven even-numbered items (ATE2, ATE4, ATE6, ATE8, ATE10, ATE12, ATE14) are the New Ecological Paradigm’s anti-ecological statements, and the deposit stores them already reverse-scored, so resp runs opposite to the sentence printed here: a 5 on ATE12, “Humans were designed to dominate the remainder of nature”, means strong DISagreement with that statement rather than agreement. The values are left as deposited, because IRW does not harmonise response direction. The source paper never states the reversal – its Table 2 reports alpha 0.949 with positive loadings on all fifteen items, which is only obtainable from reverse-scored data, and all 105 inter-item correlations in this table are positive with no item-total correlation below +0.67; careless responding was tested as an alternative explanation and ruled out. Second, the survey was fielded to CEOs of Vietnamese manufacturing firms and was in all likelihood administered in Vietnamese – the paper does not state the language – but the only wording published anywhere in the article or its supplement is English, so the item text shown here is the authors’ own English rendering of the revised NEP items rather than what respondents read; it was transcribed from Table 2, which PLOS serves only as an image. Response options carry text at the two extremes only (1 = strongly disagree, 5 = strongly agree), because the paper labels no intermediate point.
liem_2024_cleaner_production	The survey was fielded to CEOs of Vietnamese manufacturing firms and was very likely administered in Vietnamese (the paper does not state the language), but the only published wording is English, so the item text is the authors’ English from the paper’s Table 2 rather than what respondents read. Only the two extremes of the 1-5 scale are labelled (1 = strongly disagree, 5 = strongly agree).
liem_2024_customer_pressure	The survey was fielded to CEOs of Vietnamese manufacturing firms and was very likely administered in Vietnamese (the paper does not state the language), but the only published wording is English, so the item text is the authors’ English adaptation of Chu et al. (2017), from the paper’s Table 2, rather than what respondents read. Only the two extremes of the 1-5 scale are labelled (1 = strongly disagree, 5 = strongly agree).
liem_2024_green_competitive_adv	The survey was fielded to CEOs of Vietnamese manufacturing firms and was very likely administered in Vietnamese (the paper does not state the language), but the only published wording is English, so the item text is the authors’ English adaptation of Lin and Chen’s items, from the paper’s Table 2, rather than what respondents read. Only the two extremes of the 1-5 scale are labelled (1 = strongly disagree, 5 = strongly agree).
liem_2024_perceived_benefit_cp	The survey was fielded to CEOs of Vietnamese manufacturing firms and was very likely administered in Vietnamese (the paper does not state the language), but the only published wording is English, so the item text is the authors’ English from the paper’s Table 2 rather than what respondents read. The items are printed as bare benefit names with no question stem, and only the two extremes of the 1-5 scale are labelled (1 = strongly disagree, 5 = strongly agree).
liem_2024_perceived_benefit_ema	The survey was fielded to CEOs of Vietnamese manufacturing firms and was very likely administered in Vietnamese (the paper does not state the language), but the only published wording is English, so the item text is the authors’ English from the paper’s Table 2 rather than what respondents read. Only the two extremes of the 1-5 scale are labelled (1 = strongly disagree, 5 = strongly agree).
liu_2017_ssrs_support	Item text for this table is the canonical Chinese SSRS rather than the study’s own materials – the PLOS deposit is an unlabelled SPSS file and the paper reproduces no items – and its English translation was produced by IRW, not by the study’s authors. Items 5, 6 and 7 are composite scores (a sum over five family sources, and two counts of endorsed support sources), so their resp values are totals with no per-level option labels; the response format for those three is given inside item_text.
liu_2018_gse	The study surveyed working adults in Shanghai and its General Self-Efficacy Scale reliability is cited to the Chinese version of the scale, so the questionnaire was administered in Chinese; the only wording published in the article or its supplement is the English sentence attached to each data column as an Excel comment, so the item text shown here is that English (identical to Schwarzer & Jerusalem’s published English GSE) rather than what respondents read
liu_2018_lot_r	For the three pessimism items (Optimism3, Optimism7, Optimism9) the deposited responses are already reverse-coded, so their anchors are shipped reversed – 1 = strongly agree through 5 = strongly disagree – while the other seven items carry the paper’s stated 1 = strongly disagree through 5 = strongly agree; and because the study’s Chinese wording appears nowhere in the article or its supporting information, the item text shipped is the LOT-R’s English original rather than what respondents read
liu_2018_panas	The study surveyed working adults in Shanghai and the questionnaire was administered in Chinese, but the only wording published in the article or its supporting information is the English adjective attached to each data column as an Excel comment, so the item text shown here is that English rather than what respondents read. The PANAS has no item stems, so item_text is the single adjective; no instruction text is shipped, because the paper describes the task without ever printing the instruction respondents were given.
liu_2022_fragreading_cdq	The item text is the study’s own Chinese wording, from the questionnaire in the PeerJ supplement, so it is what respondents actually read. What is inferred is which sentence goes with which item code: the questionnaire numbers its items 9.1-9.11 and the data columns are N9_1-N9_11, and that numbering is the only thing tying the two together. Checks against the live data confirm the reverse-item block (N9_6-N9_9) and the direction of the 1-5 anchors, but nothing distinguishes items within a block, so a swap of, say, N9_1 and N9_2 would not be detectable.
liu_2022_fragreading_frq	Item text is shipped for all 22 items, including 8.6 and 8.10, which the study’s own supplement marks as deleted after its confirmatory factor analysis. The IRW table carries all 22, so the wording is given for all 22; those two are also the weakest items in the data (item-rest correlations 0.204 and 0.272, against 0.375-0.613 for the other twenty), which is consistent with their counter-worded content.
liu_2022_mice_skills	The item text here is the study’s own English skill labels, taken verbatim from the header row of the deposited data file; the questionnaire was administered in Chinese and no Chinese wording is published in the paper or its supporting file. Note that each item code IS its label, so the item text adds no wording beyond the code. Only the end points of the 1-5 scale are labelled in the source (1 = very unimportant, 5 = very important); options 2-4 were never labelled and are left blank rather than filled in.
liu_2023_adherence_factors	The item and instruction text is the administered Chinese, with the authors’ own English alongside, except that the English of the factor-definitions paragraph in instructions_translated was produced by IRW (the study’s English questionnaire omits it).
liu_2023_adherence_tools	Item text is the administered Chinese, with the study authors’ own English in the translated columns. Two things are transcribed exactly as the source has them rather than corrected: item 1 reads “MMS-8”, which is the study’s own typo for the MMAS-8 adherence scale, and the first response option ships the questionnaire’s fuller wording rather than the abbreviated label stored in the data file. The authors’ English renderings are much terser than their Chinese items throughout, and were kept as they wrote them.
liu_2023_attitude	The survey was in all likelihood administered in Chinese, but the only questionnaire in the study’s supplements is the authors’ English version, so that is the text shown; the questionnaire lists the three attitude items unnumbered, so their assignment to ATT1-ATT3 follows listing order; only the scale endpoints (1 strongly disagree, 5 strongly agree) are labelled.
liu_2023_brand_trust	The survey was administered in Chinese, but neither the article nor its supplements publish the Chinese wording, so the item text shown is the authors’ own English questionnaire rather than what respondents read. The questionnaire lists the four brand-trust sentences under one heading without item codes; that heading fixes that these are BT1-BT4 (the block partition reproduces the paper’s published Cronbach’s alpha of 0.875 exactly), but the order within the block follows the questionnaire’s listing order and cannot be checked against published statistics. Only the endpoints of the 1-5 scale are labelled in any source, so options 2, 3 and 4 carry no text.
liu_2023_improve_adherence	For item 3 the data file’s stored label is truncated, so the item text shipped is the fuller wording from the study’s own administered Chinese questionnaire. The authors’ published English for that item is much terser than the Chinese it translates — “Answering the patients’ questions.” against a Chinese item that also enumerates pathogenesis, harms, medication regimen and potential side effects — so the English and Chinese columns are not the same length of text here. That terseness is the authors’ own, not an IRW abridgement.
liu_2023_perceived_control	The three item codes PBC1-PBC3 are matched to the three Perceived Behavioral Control sentences in the paper’s questionnaire annexure by listing order alone. The annexure attaches no code to any sentence, and no statistical route separates three positively worded items on one 1-5 scale, so a permutation within the construct cannot be ruled out. What is established is that these three sentences are the PBC block: that partition reproduces the paper’s published Cronbach’s alpha of 0.852. The item text is also the study’s own English wording — the survey was administered in Chinese, which the deposit does not publish. One item is shipped with the source’s own broken English unrepaired.
liu_2023_purchase_behavior	The survey was administered in Chinese to consumers of Hulunbuir agricultural products, but neither the article nor its supplements publish any Chinese wording, so the item text shown here is the authors’ own English questionnaire rather than what respondents read. The questionnaire also attaches no codes to its items: its ‘Purchase Behavior’ heading fixes that these three sentences are PB1-PB3 (the block partition reproduces the paper’s published Cronbach’s alpha, 0.863), but the order within the block follows the questionnaire’s listing order and cannot be checked against the data, since the only per-item statistics the paper publishes are keyed to the codes. Only the two endpoints of the 1-5 scale are labelled in any source, so options 2, 3 and 4 carry no text.
liu_2023_purchase_intention	The survey was administered in Chinese to consumers of Hulunbuir agricultural products — the paper says its scales were translated and back-translated by the university’s English department — but neither the article nor its supplements publish any Chinese wording, so the item text shown here is the authors’ own English questionnaire rather than what respondents read. The three sentences also carry no item codes in that questionnaire. Their block is fixed: the partition reproduces all six of the paper’s published Cronbach’s alphas, 0.809 for this one. But which sentence is PI1, PI2 or PI3 follows the questionnaire’s listing order and cannot be checked against the data, whose three items have means 3.72, 3.74 and 3.71 and SDs all 0.94-0.95. Only the two endpoints of the 1-5 scale are labelled in any source, so options 2, 3 and 4 carry no text.
liu_2023_subjective_norm	The survey was in all likelihood administered in Chinese, but the only questionnaire in the study’s supplements is the authors’ English version, so that is the text shown; the questionnaire lists the three subjective-norm items unnumbered, so their assignment to SN1-SN3 follows listing order; only the scale endpoints (1 strongly disagree, 5 strongly agree) are labelled.
liu_2023_training_freq	Item text is the administered Chinese from the study’s data file, with the authors’ own English alongside. Two things to know about the response options and the English. The third of the five frequency levels ships the administered questionnaire’s wording rather than the slightly different label stored in the data file. And the authors’ English questionnaire prints these six items in a different order than both the data file and the Chinese questionnaire — its second and third items are swapped — so the English shipped here was matched to each Chinese item by content rather than by position.
liu_2025_classroom_interaction	The questionnaire was administered in Chinese, but no Chinese wording appears anywhere in the article or its supplements, so the item text shown here is the authors’ own English rendering (S1 Appendix) rather than what respondents read. Two of the eleven items are shipped in the fuller wording the S1 Appendix gives them, which differs slightly from the abbreviated column label used in the data workbook. Response options carry text at the two extremes only (1 = strongly disagree, 5 = strongly agree); the paper labels no intermediate point.
liu_2025_foreign_lang_enjoyment	The survey was administered in Chinese to 623 Chinese undergraduates — the paper says its English items were translated into Chinese and checked by a translation specialist — but neither the article nor its supplements publish any Chinese wording, so the item text shown here is the study’s own English rather than what respondents read. Only the two endpoints of the 1-5 scale are labelled in any source, so options 2, 3 and 4 carry no text. Note that the article’s prose disagrees with the study’s own data about which items form which sub-scale. Section 2.2.4 assigns teacher appreciation to items 1-3 and personal enjoyment to items 4-6, but the data file and the S1 Appendix both put personal enjoyment at 1-3 and teacher appreciation at 4-6 — and the prose’s own examples confirm the data rather than the prose, since “The teacher is friendly.” is item 5 and “I’ve learnt interesting things.” is item 2. The item codes here follow the data. Social enjoyment (items 7-9) is unaffected.
liu_2025_learning_motivation	The questionnaire was administered in Chinese, and the Chinese item wording shown here is verbatim from the study’s own data file, with the study’s own English translation alongside. The English was matched to each Chinese item by content rather than by position, because the paper’s S2 Appendix prints the sixteen items in a different order than the administered one. The response options are the exception to all of this: no Chinese anchor wording appears anywhere in the article, its appendix or the data file, so option_text carries the appendix’s English endpoints (1 = Completely Disagree, 5 = Completely Agree) and options 2, 3 and 4 carry no text because no source labels them.
liu_2025_nlgz	The questionnaire was administered in Chinese to middle-school students in Kaifeng, China, but neither the article nor its data deposit publishes any Chinese wording, so the item text shown here is the authors’ own English rendering from their Dryad README rather than what respondents read. The five response points carry no text, because no source names an anchor for any of them; the README states only that higher numbers mean greater agreement.
liu_2025_positive_cognition	The English item wording on this table was translated from the administered Chinese by IRW, not by the study’s authors: no published English exists for the 12 items of this Chinese revision of the APNIS, whose original is paywalled. The Chinese in item_text is the study’s own administered wording, verbatim from its data deposit, and the five response anchors are the study’s own English.
liu_2025_speaking_selfefficacy	The study administered these 14 items in Chinese translation, but the administered Chinese is not published anywhere in the deposit, so the wording shipped here is the authors’ own English source version from the S1 Appendix rather than what respondents read. Only the 1 and 5 anchors of the five-point scale are labelled in any source.
liu_2025_willingness_communicate	The questionnaire was administered in Chinese to 623 Chinese undergraduates — the paper says the English scale items were translated into Chinese and checked by a translation specialist — but neither the article nor its two supplements publish any Chinese wording, so the item text shown here is the authors’ own English questionnaire rather than what respondents read. Only the two endpoints of the 1-5 scale are labelled in any source, so options 2, 3 and 4 carry no text.
liu_2025_ydcy	The deposit’s README describes this block as the Godin Leisure Time Exercise Questionnaire, but the data are the PARS-3: the file’s own composite column equals YDCY2 x YDCY3 x YDCY4 exactly for all 879 respondents, while the README’s Godin weighting reproduces 5 of 879. The item text shown here is therefore PARS-3 wording rather than the README’s instrument. YDCY3 is the duration item, identifiable as the only one scored from 0: its lowest level (“under 10 minutes”) is stored as 0, so YDCY3 runs 0-4 where the other PARS-3 items run 1-5. Intensity and frequency (YDCY2, YDCY4) cannot be told apart from these data, because the composite is a product and so is symmetric in them; their assignment follows the canonical PARS-3 order. The survey was administered in Chinese and no Chinese wording is published anywhere in the deposit or the paper, so the English shown is a published English form of the same instrument rather than what respondents read.
lorenz_2016_efficacy1	This study was run in German. The item text is the German General Self-Efficacy wording respondents read — three items verbatim from the paper’s own S1 Appendix and the other seven from the scale authors’ own page — with the instrument publisher’s English alongside. The German and English versions of this scale are published in different item orders, so the English is a parallel version rather than a translation of the German sentence beside it. The study used a 6-point format instead of the scale’s own 4-point one and publishes English labels only for its endpoints, so options 2 to 5 carry no text. Which self-efficacy item each of the codes efficacy1.1 to efficacy1.10 refers to is reconstructed rather than stated by the source. The reconstruction reproduces the paper’s published confirmatory factor analysis exactly, but items whose published means are tied are not individually distinguished by it.
lorenz_2016_hope	This study was run in German, but only half of its German wording was ever published. The paper’s appendix prints the German for the three items that entered its CPC-12 analysis (hope1, hope4, hope5), and those ship as respondents read them with English alongside; for hope2, hope3 and hope6 no German appears in the paper or its supplements, so what is shown is the State Hope Scale’s original English rather than the administered sentence. The study also used a 6-point format in place of the scale’s original 8-point one and publishes English labels only for the two endpoints, so options 2 to 5 carry no text. Which State Hope Scale item each code refers to is reconstructed rather than stated by the source. hope1, hope3, hope4 and hope5 are each pinned individually by reproducing the paper’s published factor loadings exactly, but hope2 and hope6 are both agency items outside that analysis and are not distinguished from one another.
lorenz_2016_optimism2	This study was run in German. The item text is the German LOT-R wording respondents read, from the published German version the paper cites (Glaesmer et al. 2008), with the scale authors’ own English alongside. The German and English LOT-R are parallel versions rather than sentence-for-sentence translations, so item 5 reads “I feel comfortable among my friends” in German where the English reads “I enjoy my friends a lot”. Two things about the response values. The three pessimism items (optimism2.3, .7, .9) are stored already reverse-coded, so a 6 on those means the respondent chose “strongly disagree”, and the option labels shipped for them run the opposite way from every other item. That was established from the data, not from any statement in the deposit: the six scored items as stored reproduce the paper’s published alpha of .74, whereas recoding them as the LOT-R manual directs gives an alpha of -0.02. Separately, the study replaced the LOT-R’s own 5-point format with a 6-point one and published English endpoint labels only, so options 2 to 5 carry no text. Which LOT-R item each code refers to is reconstructed. The reconstruction fixes the four filler items and the optimism/pessimism split and reproduces the paper’s published factor analysis exactly, but optimism2.1 and optimism2.4 are not distinguished from each other, nor are the members of {optimism2.3, .7, .9} or of {optimism2.2, .6, .8}.
lshs-e_quidaja_2022	The item text here is the administered Spanish, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. Published English exists for the LSHS-E, but this administration is a Spanish adaptation with its own item set, so the administered Spanish is what is translated.
lu_2017_gad7	The GAD-7 was administered in Chinese, but no Chinese wording is published in the paper or its two supplementary data files, so the item and option text shipped here is the instrument publisher’s official English (phqscreeners.com) rather than the sentences respondents read. Note also that the response table covers the deposit’s full 1296 respondents, more than the 1096 the paper reports analysing, and that it contains repeated id-item pairs. Those are not a doubled upload: they are a second wave of measurement for 129 respondents, and the responses differ across waves for many of them.
lu_2017_phq9	The PHQ-9 was administered in Chinese, but no Chinese wording is published in the paper or its two supplementary data files, so the item and option text shipped here is the instrument publisher’s official English (phqscreeners.com) rather than the sentences respondents read. Note also that the response table covers the deposit’s full 1296 respondents, more than the 1096 the paper reports analysing, and that it contains repeated id-item pairs. Those are not a doubled upload: they are a second wave of measurement for 129 respondents.
lunacortes_2019_interperson_conn	Interviews were conducted in Spanish, but the Spanish wording is not published, so the item text is the English the authors translated from; the paper notes that some verbs were changed to simpler Spanish synonyms after pre-testing, so the Spanish was not a strict rendering of this English. The wording comes from the article’s Table 1, published only as an image, and its four bullets are unnumbered, so ic_1 to ic_4 follow bullet order. ic_1 is independently identified as the weakest indicator, but ic_2, ic_3 and ic_4 are not separated by any published statistic. The source spelling ‘holydays’ is kept.
lunacortes_2019_satisfaction	Item wording for this table comes from the source article’s Table 1, which PLOS publishes only as an image, so the three sentences were transcribed by eye rather than extracted as text. Their bullets are unnumbered, so the assignment of wording to satis_1, satis_2 and satis_3 follows bullet order. That is corroborated by the data — the observed single-factor loadings 0.78/0.72/0.66 reproduce the paper’s published rank order — which separates satis_3 firmly but leaves the satis_1/satis_2 ordering resting on a small gap. The instrument was administered in Spanish; the shipped text is the English the authors translated from, since the Spanish wording is not published anywhere in the article or its supplement. Only the two endpoints of the 1-7 scale are labelled in the source, so options 2 to 6 carry no text.
lunacortes_2019_self_congruity	Interviews were conducted in Spanish, but the Spanish wording is not published, so the item text is the English the authors translated from; the paper notes that some verbs were changed to simpler Spanish synonyms after pre-testing. The wording comes from the article’s Table 1, published only as an image, and its four bullets are unnumbered, so self_cong1 to self_cong4 follow bullet order. The published loadings separate items 1-2 from items 3-4 and the data reproduce that split, but the order within each pair is not checked. Item 4 keeps the source’s ‘how I like be seen by others’.
lunacortes_2019_social_value	Item wording for this table comes from the source article’s Table 1, which PLOS publishes only as an image, so the three sentences were transcribed by eye. Their bullets are unnumbered, so the assignment of wording to social_val1, social_val2 and social_val3 follows bullet order. social_val3 is pinned independently: the paper dropped its third social-value item from the confirmatory analysis, and its published two-item alpha of 0.822 is reproduced only by the pair social_val1 + social_val2 (0.8219, against 0.34 and 0.46 for the other pairs); social_val3 is also the weakest indicator in the data. The social_val1 vs social_val2 ordering is not pinned, since the paper’s loadings for those two share an identical robust t. The instrument was administered in Spanish; the shipped text is the English the authors translated from, as the Spanish is not published.
luo_2021_acculturation_index	The item text is the study’s SPSS column names, with word spacing restored by IRW. Two names were cut off at SPSS’s 64-character limit: the last word of item_01 comes from the paper’s own quotation, and the word ‘country’ ending item_10 was completed by IRW (anything after it is lost). item_06 follows the paper’s quotation ‘I should remain attached to my original culture’, where the column name omits ‘remain’. Despite the table name, the measure is the study’s adaptation of Swaidan et al.’s Adopt and Keep scale (items 1-5 adopt, 6-10 keep), not Ward and Kennedy’s Acculturation Index.
luo_2021_conational_ties	This instrument’s wording survives only in the deposit’s SPSS column names, which SPSS caps at 64 characters and strips of spaces. All four are cut off mid-word, so only the first 48 to 61 characters of each shipped question are the study’s own text; the remainder of each sentence, together with all punctuation and word spacing, was completed from the same four Multi-Dimensional Support Scale questions as printed in a different paper (Wei 2025, Front Psychol 16:1607241). Which of the four questions each item code refers to is nonetheless exact: each item’s live response distribution matches its source column cell for cell, and only 1 of the 24 possible orderings fits. Scale points 2 and 3 carry no option text, because the study labels only 1 = never and 4 = always.
ly_2021_animal_empathy	Each item is one emotion rating for one 10-second video clip; the item text is the emotion name and the clip is described in the section prompt. The study published a single interviewer blurb per video code, and each code covers two different exemplar clips shown to different halves of the sample (CB1 and CB2, for instance), so those pairs carry identical section_prompt text even though respondents saw different videos.
lys_2020_rape_1_rma	The item text here is the administered Polish, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors.
lys_2020_rape_2_rma	The item text here is the administered Polish, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors.
lys_2020_rape_2_rma_pre	The item text here is the administered Polish, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors.
lys_2020_rape_3_asi	The item text here is the administered Polish, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. Glick & Fiske’s published English for the Ambivalent Sexism Inventory exists and the item count matches, but the correspondence is not clean enough to rely on – the Polish carries a short form where the original pairs a reverse-scored item with a long one, so using the original’s numbering could invert an item’s polarity.
lys_2020_rape_3_diff	Administered in Polish; the item text is the study’s own English rendering from its SPSS labels, since the Polish originals are in neither the deposit nor the paper. The 13 items are not one scale: they form two opposed subscales, belief in biological and in social origins of gender differences, which the deposit scores separately. Do not sum all 13.
lys_2020_rape_3_kpnts	The English in item_text_translated is a MACHINE TRANSLATION produced by IRW from the administered Polish – it is not the instrument’s own English and has not been reviewed by a Polish speaker. item_text is the administered Polish, which is authoritative; where the two diverge the translation follows the Polish, so kpnts7 and kpnts19 read ‘better’ / ‘worse’ groups rather than the canonical scale’s ‘superior’ / ‘inferior’. Two further things to know about this table. The item codes are not contiguous – kpnts2, 3, 5, 7, 8, 10, 11, 13, 14, 16, 17, 19, 20, 22, 23, 24, sixteen items drawn from a longer block – and the eight reverse-worded items (kpnts3, 8, 10, 14, 16, 20, 23, 24) are already recoded in the deposit, so higher resp means more social dominance orientation for every item and the deposit’s SDO score is the plain sum of all sixteen (reproduced exactly for 370 of 370 complete rows). Do not reverse them again. Finally, the code misleads: despite kpnts, this table IS the Social Dominance Scale, while lys_2020_rape_3_sds is NOT – it is the 29-item social DESIRABILITY scale.
lys_2020_rape_3_rma	The item text here is the administered Polish, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors.
lys_2020_rape_3_rwa	The Polish in item_text is NOT established as what respondents read: it comes from Radkiewicz (2011), an Altemeyer/Duckitt-lineage 20-item Polish RWA, whereas the study cites Grzesiak-Feldman’s (2012) Polish adaptation of Funke’s 12-item scale. The twelve items match in content, one appendix item each, but the Polish strings may differ. For this table the study’s own English in item_text_translated is the more authoritative record. The six odd-coded reverse-worded items are already recoded in the deposit; do not reverse them again.
lys_2020_rape_3_sj	Administered in Polish; the item text is the study’s own English rendering from its SPSS labels, since the Polish originals are in neither the deposit nor the paper. sj3 and sj7 are reverse-worded but already recoded in the deposit, so higher resp means more system justification for every item; do not reverse them again.
lys_2020_rape_3_unjust	Administered in Polish; the item text is the study’s own English rendering from its SPSS labels, since the Polish originals are in neither the deposit nor the paper.
ma2026_bsmas	The scale was administered in Chinese to primary school students, but neither the study’s figshare deposit nor its preprint publishes the Chinese wording, so the item text shown here is the canonical English rather than what respondents read. Note that item 6’s canonical reference to “your job/studies” is the original scale’s wording and not something these primary-school respondents would have been asked.
ma2026_sabas	The scale was administered in Chinese to primary school students, but neither the study’s figshare deposit nor its preprint publishes the Chinese wording, so the item text shown here is the instrument developers’ own English rather than what these respondents read.
machado_2020_cat_separation	The questionnaire was administered in Portuguese, but the only wording published in the article or its supplement is the authors’ own English version, so that is what is shown here rather than what respondents read. The translated columns are empty for the same reason: there is no second version to put in them.
machivallianism_test_vcl	This table is not the Machiavellianism scale: it is the 16-word vocabulary check-list administered in the same openpsychometrics session as the MACH-IV, in which respondents ticked every word whose definition they were sure they knew (1 = checked, 0 = unchecked). VCL6 “cuivocal”, VCL9 “florted” and VCL12 “verdid” are deliberate non-words the study included as a validity check, not transcription errors, and should not be corrected to real words. correct_response is blank for all 16 items because the source publishes no scoring key.
maes_2020_ospaq	Respondents completed a Dutch-language OSPAQ that is not published. The instructions shown are the authors’ English rendering from the paper, and the item text is the study’s English data-file column labels for the four postures (with the typo ‘demading’ corrected), not the questionnaire wording or the canonical OSPAQ category names. The study asked about a typical working day rather than the OSPAQ’s usual last seven days. Responses are free-entry percentages, so there is no option text.
magiccats_ozono_2020	The participant file stores both ratings starting at 0 (interest 0-9, performing 0-2), one point below the 1-10 and 1-3 scales the paper describes, so the published anchors are shipped at stored value = described value minus 1. The item text is the data file’s short column label; the on-screen question wording was not published.
makai_2023_entrepreneurial_transdanubia	Item text for this table is the English wording published in the study’s own S1 File. The survey was administered to Hungarian students and the Hungarian original is not published anywhere in the paper or its supplement, so no administered-language text is available. One item code carries a typo present in the source file (“cfareer”).
makkiyah_2023_eq	As for makkiyah_2023_srl: IRW’s English translation of the Indonesian headers.
makkiyah_2023_gse	As for makkiyah_2023_srl: IRW’s English translation of the Indonesian headers.
makkiyah_2023_motivation	As for makkiyah_2023_srl: IRW’s English translation of the Indonesian headers.
makkiyah_2023_srl	The English item text (item_text_translated) was written by IRW from the deposit’s Indonesian headers; it is not the authors’ wording.
makowska_2023_pdts	The paper numbers this scale’s items differently from the data file. Its Table 2 lists the “irritated” item as item 2 and the “no control over ICT changes” item as item 3, while the deposited SPSS files label the columns the other way round — and those labels carry the item numbers inside the label text itself (PDTS2. How often have you felt that you had no control…). IRW follows the data file, which is the direct tie to the codes in this table, and the response data weakly corroborates it: the undisputed “upset” item correlates 0.783 with PDTS3 against 0.733 with PDTS2 in the paper’s Study 1, and 0.660 against 0.611 in Study 2, as expected if PDTS3 is the near-synonymous “irritated” item. Readers comparing against the paper’s item numbers should be aware that PDTS2 and PDTS3 may be interchanged; the other four items are unambiguous. Separately, the item text and response options are shipped in the administered Polish. The English in the translated columns for the response options was written by IRW, because the study’s own English renderings of those anchors are inconsistent across its files.
makransky_2016_mcq_correct	This table’s resp is a 0/1 score, not a chosen option, and for MCQ7 the score and the published answer key disagree. The paper’s S1 Table marks “the differential agent” as correct, but the deposited data score the alternative “a carbon source”: every one of the 60 pre-test and 56 post-test responses scored 1 chose that option, and the counts match exactly, so the disagreement is systematic rather than a few coding slips. The published key is the scientifically correct one — in mannitol salt agar the salt is the selective agent and mannitol is the differential agent — so this is best read as a scoring error in the study’s own data. correct_response therefore carries the published key, while resp faithfully reproduces the study’s scoring. The practical consequence: for MCQ7 only, resp = 1 means the respondent chose “a carbon source”, not the answer named in correct_response.
makransky_2016_self_efficacy	The response anchors shipped here are the ones in the study’s own data file and S1 Table (“Completely disagree” to “Completely agree”), which differ in wording from the article body’s description of the same scale as running “Strongly disagree” to “Strongly agree”. The two level-1 sources agree with each other, so the article’s phrasing is the outlier.
malik_2018_individual_motivation	Item text is the study’s own terse SPSS variable labels (for example “Interest in job”, “Availability of adequate food”) rather than the full sentences respondents read. The questionnaire’s own item phrasing and its instructions are not published in the article or its supplement, so the short labels are all that exists.
malik_2018_organizational_motivation	As with its sibling tables, the item text is the study’s terse SPSS variable labels rather than the sentences respondents read. Two caveats specific to this table. One item (“Career advancement and growth”) carries no response-option labels in the source file, so the 1-5 anchors shared by the other 35 items in the same block were applied to it — an inference, not a transcription. And two typos are shipped verbatim as the source has them: “Availability of proper of proper dressing rooms”, and a double space in “Number of staff available during work”.
malik_2018_social_motivation	The source questionnaire’s socio-cultural section has 15 items but this table carries 14. The processing script treated the data file’s SC3 and CS4 as the same variable when they are two different items with different wording and different responses, so SC3 (“I am satisfied with my personal life issues”) is absent from IRW altogether, and every code from the third onward is shifted: social_3 carries CS4’s wording (“I am satisfied with the team work around me during work”). The item text shown is faithful to the data as it currently stands — each code is labelled with the column genuinely behind it. The missing item is a defect in the response data, tracked separately, and fixing it will renumber these codes. Item text here is also the study’s terse SPSS variable labels rather than the sentences respondents read.
malinowska_2021_saq_nurses	Item text is the Polish SAQ-SF PL as administered, with English alongside; the English item stems and response anchors are the instrument’s own published English rather than a translation of the Polish, while the instructions, the section prompt for items 24-28 and the ‘ward manager’/‘hospital director’ labels were translated by IRW. Items 24-28 were each rated twice, for the ward manager (kier_24a..kier_28a) and for the hospital director (bezp_24..bezp_28); the raw file labels neither, so that assignment is inferred and is documented in the table’s verification record.
malinowska_2021_saq_physicians	resp = 6 is not a sixth agreement level: it is the questionnaire’s “Nie dotyczy” (not applicable) option, which sits outside the 1-5 “Zdecydowanie nie zgadzam sie”..”Zdecydowanie sie zgadzam” scale and should be treated as missing rather than as the top of the scale – 1.7% of responses overall, but 122 of 738 on bezp_35, the pharmacist-collaboration item. Questions 24-28 were each rated twice, for the ward manager and for the hospital director; kier_24a-28a is read as the ward-manager rating and bezp_24-28 as the hospital-director rating, and the bracketed referent in item_text ([kier. oddzialu] / [dyr. szpitala]) is an editorial addition rather than part of the printed stem. The English for those ten items’ section prompt was translated by IRW; every other English field is the University of Texas SAQ-SF’s own wording.
mancone_2024_ravlt_recall	The five items are the five successive immediate-recall trials of one 15-word list, not five different questions, so item_text is blank by design and resp is the raw count of words correctly recalled on that trial (0-15); the study publishes neither the 15-word stimulus list nor a participant-facing instruction script, so instructions holds the paper’s own third-person description of the administration
manolika_2021_mini_ipip	Ten item codes ending in R (IPIP_06R … IPIP_20R) are stored already reverse-scored, although the source file labels every item 1=Strongly Disagree to 5=Strongly Agree; the option_text for those items is therefore reversed (1=Strongly Agree … 5=Strongly Disagree), and the English wording stands in for the Greek translation the study administered.
manolika_2021_movie_preferences	Respondents rated the genres in Greek; the genre names shown are the study’s own English renderings, and the Greek wording was not published.
manolika_2021_reading_preferences	Administered in Greek; the genre names and anchors shown are the study’s own English, from the paper appendix and the data file value labels, as the Greek wording was not released.
marcussonclavertz_2019_velten	The item text is the study’s own variable labels from its data supplement, with a survey-software artifact prefix stripped; the item codes are the source column names, so text and code are tied at the source with no inference. Note that the instructions field is not the instruction respondents read: the study does not publish that, so what is shown is the paper’s own third-person description of the task from its Methods section.
marquessanchez_2023_kidmed	For item ALIDES the response coding is the complement of the item as worded: the KIDMED item reads ‘No desayuna’ / ‘Skips breakfast’, but in this table resp=1 marks a respondent who does NOT skip breakfast (it adds +1 to the study’s own KIDMED total and is absent from the study’s negative-item composite, and 79.6% of respondents sit at 1). The shipped option_text is flipped for that item alone; the other 15 items are coded resp=1 = ‘Sí’
martindelcampo_2020_cvcv	Respondents answered the Spanish questionnaire; the English item wording shown is the study’s own SPSS variable labels, which are a literal and in places ungrammatical rendering (e.g. ‘in the places that I have frequently shot someone’ meaning someone was shot in places the respondent frequents), and the Spanish wording was not published.
martindelcampo_2020_tipi	Respondents answered a Spanish version of the TIPI; the English trait pairs shown are the study’s own SPSS variable labels, which differ from Gosling’s published English for six items (e.g. ‘Critic, fighter’ for ‘Critical, quarrelsome’, ‘Quiet, emotionally stable’ for ‘Calm, emotionally stable’), and the Spanish wording was not published.
mascherini_2021_meddiet	Each of the 11 food-frequency items was administered twice, once for habits before the COVID-19 lockdown (wave 1) and once for habits during it (wave 2); the two administrations differ only in tense, and the past-tense wave-1 wording is what ships. The study was run in Italian but publishes only the authors’ own English version of the questionnaire, so item_text and option_text are that English rather than what respondents read. resp is the MedDietScore, not the raw frequency: for red meat, white meat, dairy products and alcohol a HIGH resp means LOW consumption, and the option_text on those four items is mapped accordingly.
matarboumosleh_2017_gad2	The column headers in the study’s data file are terse abbreviations rather than administered wording, so while they pin which item is which, the words shown here are the official GAD-7 form’s own published item text rather than anything this study printed. The response-option wording is likewise the official form’s, because the data file carries no value labels and the paper prints no anchors.
matarboumosleh_2017_phq2	As with its sibling table, the study’s data file carries only terse column headers rather than administered wording, so the item and response-option text shown here is the official PHQ form’s own published wording rather than anything this study printed.
matarboumosleh_2017_spai26	Note which instrument this is. “SPAI” here is the 26-item Smartphone Addiction Inventory (Lin et al. 2014), not the Social Phobia and Anxiety Inventory, which is a different 45-item instrument that shares the abbreviation. All 26 items are smartphone-use content. The item wording comes from the paper’s Table 2, which is published only as an image.
matosaslopez_2022_bars_teaching_blended	The behavioural anchors (option_text) are the authors’ published English; respondents at a Spanish university answered a Spanish version whose anchor wording was not published, so only the dimension names (item_text) are the administered Spanish.
matranga_2019_hpv_knowledge	Response code 6 is not one of the five printed options (Yes, Much, Somewhat, Little, No) and its meaning is undocumented by the source, so it has no option text; the paper’s own percentages suggest it marks a non-response or don’t-know answer. The item wording is the English version printed in the paper; the study, run in Sicily, was presumably administered in Italian.
mboya_2020_gds15	Responses are stored pre-scored (1 = the depressive answer), so for the five positively worded items (satisfy0, spirt0, hapy0, alive0, energy0) resp 1 means the respondent answered ‘no’; the source data file does not label those five items’ values, and this direction was established from the data. The item wording is the English of the study’s data file; the language the interviews were conducted in is not stated in any accessible source.
meco_l1_comprehension	Item text is published for wave-1 sites only: 317 of 708 items (all wave-2-only versions: Basque, Brazilian Portuguese, Chinese, Danish, Icelandic, Serbian, Hindi, Chilean Spanish, German wave 2 and Russian wave 2) carry the reading passage but no question wording, because wave 2 never released its questions; the 20 shared matched-text items show the English wording although they were answered in 18-20 languages (language lists them per item); and English item_text_translated is a machine translation by IRW.
mede_2025_clim_emotions	The item text is the English core questionnaire of the TISP study. Most respondents (68 countries, 37 languages) answered a local translation, which is published per sample in the OSF deposit (05_survey-materials/questionnaire/countries/) but not reproduced here.
mede_2025_clim_polsupport	The item text is the English core questionnaire of the TISP study. Most respondents (68 countries, 37 languages) answered a local translation, published per sample in the OSF deposit but not reproduced here. Respondents were also offered ‘Not applicable’ (4) for every policy; those answers were set to missing in this table, so resp covers only 1-3.
mede_2025_clim_weather_future	The item text is the English core questionnaire of the TISP study. Most respondents (68 countries, 37 languages) answered a local translation, which is published per sample in the OSF deposit but not reproduced here. The instructions include the block’s lead-in sentence (‘please think about your country’), which the questionnaire prints once above the paired past-impact question.
mede_2025_clim_weather_past	The item text is the English core questionnaire of the TISP study. Most respondents (68 countries, 37 languages) answered a local translation, which is published per sample in the OSF deposit but not reproduced here.
mede_2025_goals_priority	Item text is the TISP English core questionnaire; most respondents (68 countries, 37 languages) answered a local translation, which is published per sample in the OSF deposit (05_survey-materials/questionnaire/countries/) but not reproduced here
mede_2025_goals_tackle	Item text is the TISP English core questionnaire; most respondents (68 countries, 37 languages) answered a local translation, which is published per sample in the OSF deposit (05_survey-materials/questionnaire/countries/) but not reproduced here
mede_2025_normperc	Item text is the TISP English core questionnaire; most respondents (68 countries, 37 languages) answered a local translation, which is published per sample in the OSF deposit (05_survey-materials/questionnaire/countries/) but not reproduced here. The 1-5 agreement ladder was administered with only its endpoints labelled, so no option text is shipped for 2-4
mede_2025_outspokenness	Item text is the TISP English core questionnaire; most respondents (68 countries, 37 languages) answered a local translation, which is published per sample in the OSF deposit (05_survey-materials/questionnaire/countries/) but not reproduced here. The 1-5 agreement ladder was administered with only its endpoints labelled, so no option text is shipped for 2-4
mede_2025_sciengage	Item text is the TISP English core questionnaire; most respondents (68 countries, 37 languages) answered a local translation, which is published per sample in the OSF deposit (05_survey-materials/questionnaire/countries/) but not reproduced here
mede_2025_sciinfo	Item text is the TISP English core questionnaire; most respondents (68 countries, 37 languages) answered a local translation, which is published per sample in the OSF deposit (05_survey-materials/questionnaire/countries/) but is not reproduced here
mede_2025_scipop	Item text is the TISP English core questionnaire; most respondents (68 countries, 37 languages) answered a local translation, which is published per sample in the OSF deposit (05_survey-materials/questionnaire/countries/) but is not reproduced here. The 1-5 agreement ladder was administered with only its endpoints labelled, so no option text is shipped for 2-4
mede_2025_sdo	Item text is the TISP English core questionnaire; most respondents (68 countries, 37 languages) answered a local translation, which is published per sample in the OSF deposit (05_survey-materials/questionnaire/countries/) but is not reproduced here. The 1-10 oppose/favour ladder was administered with only its endpoints labelled, so no option text is shipped for 2-9
mede_2025_trust_scientists	Item text is the TISP English core questionnaire; most respondents (68 countries, 37 languages) answered a local translation, which is published per sample in the OSF deposit (05_survey-materials/questionnaire/countries/) but is not reproduced here
mede_2025_willvul	Item text is the English core questionnaire; the TISP survey was administered in 37 languages across 68 countries, so most respondents read a translation published per sample in the OSF deposit rather than the wording shown here.
medvedev_2018_oxh	Item wording for this table is the canonical Oxford Happiness Questionnaire (Hills & Argyle 2002); the study’s own supplement labels its columns only OXH1..OXH29 and prints no item text. The alignment of item code to item number is corroborated by the data rather than stated by the source: the twelve items the data show as reverse-keyed are exactly the OHQ’s canonical reverse-scored set, the reverse-scored scale mean (4.18, SD 0.63, alpha 0.90) reproduces the paper’s published values, and oxh_2 is the one weak item the paper singles out. Order within each keying direction is not independently pinned.
medvedev_2018_pan	The Medvedev (2018) PeerJ deposit labels its PANAS columns only PAN1..PAN20 and the article reproduces just three of the twenty adjectives as examples, so the item text shown here is the canonical Watson, Clark & Tellegen (1988) PANAS wording assigned by item number rather than wording transcribed from this study’s own materials. The assignment is confirmed to the level of valence: the study’s own PANPOS and PANNEG composite columns reproduce exactly as raw sums of the canonical positive and negative item positions for all 172-173 scored respondents, so no adjective is on the wrong subscale, but nothing in the source distinguishes one adjective from another within a subscale. The instructions respondents were given are not published anywhere in the source and are left blank.
megart_tonkovic_2021	The Croatian instructions shipped for this table are the task description from the study’s informed-consent form, not the on-screen instructions participants read during the lexical decision task – those are locked in an encrypted E-Prime script file – and their English translation was produced by IRW; item_text repeats the stimulus letter string, which is also the item code, and is left untranslated because half the stimuli are meaningless pseudowords
mekheimer_2026_flp	The instrument offers five anchors (Beginner through Expert), but no respondent selected Beginner, so the live response set runs 2-5 and only four option rows ship for each item. The Beginner anchor is absent from this item table by design rather than by omission.
meloni_2015_child_disab_knowledge	The 16 statements were administered in Italian to Italian schoolchildren, but neither the PLOS ONE article nor its supplementary files publish the Italian wording, so the item text shown is the authors’ own English version from the S2 File codebook rather than what the children heard; the codebook lists the statements in two unnumbered lists, so their order within each block of item codes is inferred rather than stated.
meloni_2015_child_ia_frequency	The interviews were conducted in Italian with Italian primary-school children, but neither the article nor its supplements publish any Italian wording, so the item text shown here is the English description of each activity given in the study’s own S2 File codebook; the activities were administered as pictures rather than as printed sentences, and the within-block order of the item codes was inferred from that codebook’s list order
meloni_2015_child_ia_satisfaction	The interviews were conducted in Italian with Italian primary-school children, but neither the article nor its supplements publish any Italian wording, so the item text shown here is the English description of each activity given in the study’s own S2 File codebook; the activities were administered as pictures rather than as printed sentences, the within-block order of the item codes was inferred from that codebook’s list order, and note that the instructions quoted here cover the whole task and so also name the Never/Once-a-year/Once-a-week card, which scores the sibling table meloni_2015_child_ia_frequency – this table’s 1-5 is the preference ordering, 1 = liked the least, 5 = liked the most
meloni_2015_deq_ce_child	The interviews were conducted in Italian with Italian primary-school children, but neither the article nor its supplements publish any Italian wording, so the item text shown here is the English of the study’s own S2 File codebook; that codebook prints only the masculine form of the eleven statements, so the items for the two female stimuli (Maria and Elena) also show the masculine wording rather than the feminine wording the children actually heard, and the statement numbering DEQ1..DEQ11 follows the codebook’s list order, which the data corroborate but do not fully pin
meloni_2015_deq_ce_parent	The parent questionnaire was administered in Italian, and neither the article nor its supplements publish any Italian wording, so the item text shown here is the English given in the study’s own S2 File codebook; the codebook also prints only the masculine form of the eleven statements, although two of the four stimuli (Maria, blind; Elena, able-bodied) are female and were administered in feminine form, so the masculine wording is shown for all 44 items.
meloni_2015_deq_oe_child	The 36 “items” here are not questions the children answered. Each child gave one spoken free-text explanation for each of four stimuli (a man in a wheelchair, a blind woman, an autistic child, an able-bodied woman), and two coders blind to the hypotheses then counted how many expressions in that answer fell into each of nine disability-model categories, so the response is a count of coded mentions rather than a chosen option: the item text is the S2 File codebook’s own verbatim coding criterion for the category being counted, the question actually put to the child is given as the section prompt, and no response-option wording exists to show. The interviews were conducted in Italian with Italian primary-school children and neither the article nor its supplements publish any Italian wording, so everything shown here is the English of the study’s own codebook. The grouping of the nine categories into the individual, social and biopsychosocial models, and the identity of the able-bodied stimulus, are pinned by the article’s Fig 1; the order within the ethical/aesthetic/religious group and the separation of the wheelchair and blind stimuli follow the codebook’s list order and the categories’ own content, which the data corroborate but do not fully pin.
meloni_2015_deq_oe_parent	As for the sibling table meloni_2015_deq_oe_child, the 36 “items” are not administered questions: parents wrote a free-text answer to each of four stimuli (a man in a wheelchair, a blind woman, an autistic child, an able-bodied woman) and two blind coders then assigned those answers to nine disability-model categories, so the response is a count of coded mentions, the item text names the coding category, the question actually asked is in the section prompt, and there is no response-option wording. This table shows the bare category heading (“Ethical Model”) where the child table shows the codebook’s fuller coding criterion for the same category; both are the source’s own text. The questionnaire was administered in Italian and neither the article nor its supplements publish any Italian wording, so the text shown is the English given in the study’s own S2 File codebook. The macro-model grouping and the able-bodied stimulus are pinned by the article’s Fig 1 to within 0.008; the order within the individual and social groups, and the separation of the three disability stimuli, rest on the categories’ content rather than a published per-condition number.
meloni_2015_parent_divers_ed	The questionnaire was administered in Italian and neither the article nor its supplements publish any Italian wording, so the item text shown here is the English given in the study’s own S2 File codebook. Each of the five sections corresponds to a photograph shown to the parent – a person with a physical disability, a person with Down’s syndrome, a boatload of migrants, two gay men kissing, a pelican covered in oil – and section_prompt holds the study’s own description of that image, not a caption respondents read; the twelve statements were repeated in randomised order under every image.
meloni_2015_parent_interests	The questionnaire was administered in Italian to Italian parents, but neither the PLOS ONE article nor its two supporting files publish any Italian wording, so the item text shown here is the authors’ own English from the S2 File codebook rather than what the parents read; the codebook lists the 25 statements as one unnumbered roman-numbered run tied to no variable code, so the order of the items within each conceptual block (music, religious, sport, cultural, social) is inferred from that list order. The response labels are also partly IRW’s: the codebook says only ‘Scale: Important/Not Important from 1 to 5’ and never labels a scale point, so ‘Not Important’ is shown at 1 and ‘Important’ at 5 - a placement supported by the data (reading books averages 4.46 and watching sport on television 1.96) - and points 2 to 4 are deliberately left blank.
menaldi_2023_brief_cope	The English in option_text_translated was written by IRW: the study administered its own Indonesian frequency anchors (Tidak pernah / Kadang-kadang / Cukup sering / Sangat sering) rather than the Brief COPE’s published anchors, so no authors’ English exists for them. item_text_translated is Carver’s original English for the correspondingly numbered item, a parallel version rather than a translation of the Indonesian wording.
mendes_2019_snycq	The Short-NYC-Q was administered at MPI-CBS Leipzig to native German speakers, but no German wording is published in the deposit or the paper, so the item text shown here is the English wording from the deposit’s own SNYCQ.json data dictionary; responses are a continuous 0-100 visual-analogue scale, so only the two endpoints carry option text
meng_2017_referent_assignment	Item codes EQ and EQ2 each cover two literal wordings under the study’s counterbalanced design – the name question in name-first (NC) trials and the color question in color-first (CN) trials (the trial’s order is in cov_pattern) – so both are given in item_text with the condition named in brackets; AQ, AQ2 and AQ3 share one identical question (“What about this?”) and are distinguished only by their position in the fixed EQ/AQ/EQ2/AQ2/AQ3 sequence, not by wording.
mental_health_wang_2024_gad	Responses are coded 1-4 rather than the GAD-7’s usual 0-3, and the deposit does not label them; the anchors shown (1=Not at all … 4=Nearly every day) are assigned in the standard ascending order, and the English wording stands in for the (presumably Chinese) version the study administered.
mental_health_wang_2024_phq	Respondents (Chinese college students) most likely answered a Chinese-language PHQ-9 that the deposit does not publish; the item text is the deposit’s own English column headers, the instructions and response labels are the canonical English PHQ-9, and resp 1-4 are read as the standard 0-3 scoring shifted up by one (1 = Not at all, 4 = Nearly every day), an inference the deposit does not state.
merlo2025_eet	The questionnaire was administered in Italian; item_text and instructions are the administered Italian verbatim from the deposit’s own data dictionary, and the English in the *_translated columns was produced by IRW rather than by the authors. The Italian response anchors were never published, so option_text instead carries the authors’ English (never/rarely/sometimes/often/almost always/always, coded 1-6) taken from the paper; the direction is confirmed by the questionnaire’s own attention-check item EET_17, whose wording is “Seleziona la risposta Mai” and which 1034 of 1065 respondents answered with resp = 1. Note that EET_17 is that attention check and not a food, so it should be dropped from any diet score pooling the other 28 items.
merlo2025_eng_cognitive	The questionnaire was administered in Italian; item_text and instructions are the administered Italian verbatim from the deposit’s own data dictionary, and the English in the *_translated columns was produced by IRW rather than by the authors. The Italian response anchors were never published, so option_text instead carries the authors’ English for the two ends of the scale only - Never at resp = 1 and Always at resp = 7, per Merlo et al. sec. 3.2.6 and Mameli & Passini (2017) - and the five interior scale points are left blank because no source labels them. The direction is confirmed numerically: rebuilding the paper’s own ENG_C composite from the live data gives mean 4.229 and skew -0.318 against a published 4.229 and -0.317, where the reversed coding would give 3.771 and +0.318. Note also that the deposit’s data dictionary has had its commas and colons flattened to spaces by its exporter, and the shipped Italian reproduces it as-is without restoring punctuation.
merlo2025_eng_emotional	The questionnaire was administered in Italian; item_text and instructions are the administered Italian verbatim from the deposit’s own data dictionary, and the English in the *_translated columns was produced by IRW rather than by the authors. The Italian response anchors were never published, so option_text instead carries the English coding stated in the paper (1 = completely disagree to 7 = totally agree) with the five interior points left blank because the source labels only the two ends. One item is stored reverse-scored: ENG_EMO_09 (“Penso che studiare sia noioso” / “I think studying is boring”) correlates +0.21 to +0.39 with all eight positively worded items and Cronbach alpha is 0.890 as stored against 0.820 with it flipped, so for that item alone resp 7 means “completely disagree” and resp 1 means “totally agree”; do not reverse it again before scoring.
mexico_2023_mobility_anxiety	The English item, instruction and response-option wording for this table was produced by IRW, not by the Centro de Estudios Espinosa Yglesias, which publishes the ESRU-EMOVI 2023 questionnaire in Spanish only; the administered Spanish is in the base fields. The two items are the GAD-2 anxiety symptoms, but they were asked about the past week on the survey’s own 1-4 frequency scale (1 ‘Rara vez o nunca (menos de 1 día)’ to 4 ‘Todo el tiempo o la mayoría del tiempo (5-7 días)’) rather than the GAD-2’s standard two-week, 0-3 scale. The respondent show card prints the top option as ‘(6-7 días)’; the questionnaire read aloud and the data labels say ‘(5-7 días)’, which is what ships.
mexico_2023_mobility_appliances	The English item, instruction and response-option wording for this table was produced by IRW, not by the Centro de Estudios Espinosa Yglesias, which publishes the ESRU-EMOVI 2023 questionnaire in Spanish only; the administered Spanish is in the base fields. The question asks what the respondent’s household had when the respondent was 14 years old, not now (the current-household version is mexico_2023_mobility_articles); spontaneous ‘NS’ (don’t know) answers, 18-116 per item, are dropped to missing by the IRW processing script.
mexico_2023_mobility_articles	The English item, instruction and response-option wording for this table was produced by IRW, not by the Centro de Estudios Espinosa Yglesias, which publishes the ESRU-EMOVI 2023 questionnaire in Spanish only; the administered Spanish is in the base fields. The question asks about the respondent’s current household (the age-14 version is mexico_2023_mobility_appliances); codes are 1 ‘Sí’, 2 ‘No’.
mexico_2023_mobility_assets	The English item, instruction and response-option wording for this table was produced by IRW, not by the Centro de Estudios Espinosa Yglesias, which publishes the ESRU-EMOVI 2023 questionnaire in Spanish only; the administered Spanish is in the base fields. The question asks whether anyone in the respondent’s household owned each property or financial product when the respondent was 14 years old; spontaneous ‘NS’ (don’t know) answers are dropped to missing by the IRW processing script, which removes 146-255 respondents per item on the property rows and 504-875 on the financial products. CEEY’s own data file labels items p32a-p32n with the stem of a different question (P.31, ‘A los 14 años, ¿tenían?’); the stem shipped here is the one printed in the questionnaire.
mexico_2023_mobility_community	The English item, instruction and response-option wording for this table was produced by IRW, not by the Centro de Estudios Espinosa Yglesias, which publishes the ESRU-EMOVI 2023 questionnaire in Spanish only; the administered Spanish is in the base fields. Spontaneous ‘No sabe’ (don’t know) answers are dropped to missing by the IRW processing script, which removes 4% of respondents on the daycare item (p98d) and 7% on the older-adult day-centre item (p98i).
mexico_2023_mobility_finances	The English item, instruction and response-option wording for this table was produced by IRW, not by the Centro de Estudios Espinosa Yglesias, which publishes the ESRU-EMOVI 2023 questionnaire in Spanish only; the administered Spanish is in the base fields. The question asks whether the respondent or their spouse or partner has each property or financial product; codes are 1 ‘Sí’, 2 ‘No’.
mexico_2023_mobility_mood	The English item, instruction and response-option wording for this table was produced by IRW, not by the Centro de Estudios Espinosa Yglesias, which publishes the ESRU-EMOVI 2023 questionnaire in Spanish only; the administered Spanish is in the base fields. The seven items are CES-D-style depressive symptoms (plus ‘sin esperanza’, which is not in the CES-D-20) asked about the past week on the survey’s own 1-4 frequency scale (1 ‘Rara vez o nunca (menos de 1 día)’ to 4 ‘Todo el tiempo o la mayoría del tiempo (5-7 días)’), not the CES-D’s 0-3 coding. p107g (‘disfrutó de la vida’) is positively worded and stored as recorded, so on that item alone a higher value means better mood. The respondent show card prints the top option as ‘(6-7 días)’; the questionnaire read aloud and the data labels say ‘(5-7 días)’, which is what ships.
mexico_2023_mobility_necessities	The English item, instruction and response-option wording for this table was produced by IRW, not by the Centro de Estudios Espinosa Yglesias, which publishes the ESRU-EMOVI 2023 questionnaire in Spanish only; the administered Spanish is in the base fields. Items p105l-p105o (children’s shoes, play equipment, a lit desk, an annual check-up) were asked only in households with a member aged 0-17, so they have about 7,900 respondents against about 17,800 for the other items; spontaneous don’t-know answers (7-173 per item) are dropped to missing by the IRW processing script.
mexico_2023_mobility_neighborhood	The English item, instruction and response-option wording for this table was produced by IRW, not by the Centro de Estudios Espinosa Yglesias, which publishes the ESRU-EMOVI 2023 questionnaire in Spanish only; the administered Spanish is in the base fields. The question asks about public services in the neighbourhood where the respondent lived at age 14 (the current-neighbourhood version is mexico_2023_mobility_community); spontaneous ‘NS’ (don’t know) answers are dropped to missing by the IRW processing script, which removes 4% of respondents on the daycare item (p33d) and 6% on the older-adult day-centre item (p33i).
mexico_2023_mobility_rooms	The English item, instruction and response-option wording for this table was produced by IRW, not by the Centro de Estudios Espinosa Yglesias, which publishes the ESRU-EMOVI 2023 questionnaire in Spanish only; the administered Spanish is in the base fields. The question asks about the dwelling where the respondent lived at age 14 (the current-dwelling version is mexico_2023_mobility_spaces); spontaneous ‘NS’ (don’t know) answers, under 0.4% per item, are dropped to missing by the IRW processing script.
mexico_2023_mobility_services	The English item, instruction and response-option wording for this table was produced by IRW, not by the Centro de Estudios Espinosa Yglesias, which publishes the ESRU-EMOVI 2023 questionnaire in Spanish only; the administered Spanish is in the base fields. The question asks about the respondent’s current dwelling (the age-14 version is mexico_2023_mobility_utilities); codes are 1 ‘Sí’, 2 ‘No’.
mexico_2023_mobility_spaces	The English item, instruction and response-option wording for this table was produced by IRW, not by the Centro de Estudios Espinosa Yglesias, which publishes the ESRU-EMOVI 2023 questionnaire in Spanish only; the administered Spanish is in the base fields. The question asks about the respondent’s current dwelling (the age-14 version is mexico_2023_mobility_rooms); codes are 1 ‘Sí’, 2 ‘No’.
mexico_2023_mobility_utilities	The English item, instruction and response-option wording for this table was produced by IRW, not by the Centro de Estudios Espinosa Yglesias, which publishes the ESRU-EMOVI 2023 questionnaire in Spanish only; the administered Spanish is in the base fields. The question asks about the dwelling where the respondent lived at age 14 (the current-dwelling version is mexico_2023_mobility_services); spontaneous ‘NS’ (don’t know) answers, under 0.4% per item, are dropped to missing by the IRW processing script. Item p26e ships the questionnaire’s wording, ‘servicio doméstico (personas que reciben un pago por limpiar, lavar, planchar; jardinero, chofer particular)’, where CEEY’s data file labels it ‘personal remunerado del hogar’.
mexico_2023_quality_administration	The item text here is the administered Spanish, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors.
mexico_2023_quality_buses	Fuente: INEGI, Encuesta Nacional de Calidad e Impacto Gubernamental (ENCIG) 2023 y ENCIG 2021 – wording transcribed from INEGI’s 2023 questionnaire (question 5.9/5.9a), English translated by IRW, response codes shifted (Si = 0, No = 1), and no INEGI endorsement is implied; rows with cov_year = 2021 answered ENCIG 2021 question 5.8/5.8a, the same bus/van/combi/microbus items asked about 2021 in non-gender-inclusive wording (e.g. ‘pasajeros’ for ‘pasajeros(as)’).
mexico_2023_quality_cablecars	Fuente: INEGI, Encuesta Nacional de Calidad e Impacto Gubernamental (ENCIG) 2023 – wording transcribed from INEGI questionnaire (question 5.11/5.11a, asked only of Cablebús/Mexicable users in Mexico City and the State of Mexico), English translated by IRW, response codes shifted (Sí = 0, No = 1), and no INEGI endorsement is implied.
mexico_2023_quality_confidence	Fuente: INEGI, Encuesta Nacional de Calidad e Impacto Gubernamental (ENCIG) 2023; IRW transcribed question 11.1, stripped interviewer directions and added an English translation of its own, and INEGI does not endorse this item text. resp 5 is ‘No aplica’ (not applicable), not a scale point. Items p11_1_10 to p11_1_25 pool the 2021 and 2023 surveys, where 2021 used slightly different wording for items 11, 17, 22, 23 and 25; items p11_1_01 to p11_1_09 hold 2023 responses only because the processing script dropped the 2021 answers (INEGI named those columns without the leading zero).
mexico_2023_quality_corruption	Fuente: INEGI, Encuesta Nacional de Calidad e Impacto Gubernamental (ENCIG) 2023 – IRW transcribed section VIII, stripped interviewer directions, shifted the response codes (Si = 0, No = 1) and added an English translation of its own, and INEGI does not endorse this item text; the table also holds ENCIG 2021 respondents (cov_year = 2021), who were asked the same questions but read ‘Durante 2021’ in the 8.3 prompt and ungendered wording throughout.
mexico_2023_quality_corruptionperception	The item text here is the administered Spanish, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. Mexican institution names are kept in Spanish with an English gloss where the acronym would otherwise be opaque (CONAPRED, INE, CNDH, INEGI).
mexico_2023_quality_drainage	The item text here is the administered Spanish, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. Item texts carry the source numbering and trailing punctuation verbatim.
mexico_2023_quality_electricgovernment	Fuente: INEGI, Encuesta Nacional de Calidad e Impacto Gubernamental (ENCIG) 2023 – IRW transcribed section X (question 10.1), stripped interviewer directions, shifted the response codes (Si = 0, No = 1) and added an English translation of its own, and INEGI does not endorse this item text; the table also holds ENCIG 2021 respondents (cov_year = 2021), who were asked the same six questions but read ‘Durante 2021’ in the prompt and ‘etc.’ for ‘etcetera’ in item 4.
mexico_2023_quality_generalcorruption	Fuente: INEGI, Encuesta Nacional de Calidad e Impacto Gubernamental (ENCIG) 2023 – IRW transcribed section IX questions 9.1 and 9.7, stripped interviewer directions, shifted the response codes (Sí = 0, No = 1; 9.7’s ‘No aplica’ kept as 3, and it is the answer of about three quarters of p9_7 rows) and added an English translation of its own, and INEGI does not endorse this item text; the table also holds ENCIG 2021 respondents (cov_year = 2021), who were asked the same questions in ungendered wording and without ‘o Fiscalía Estatal’ in 9.7’s list of authorities.
mexico_2023_quality_health	Fuente: INEGI, Encuesta Nacional de Calidad e Impacto Gubernamental (ENCIG) 2023 y ENCIG 2021 – IRW transcribed question 5.4/5.4a (health service in the IMSS) from INEGI’s 2023 questionnaire, stripped interviewer directions, shifted the response codes (Sí = 0, No = 1) and added an English translation of its own, and INEGI does not endorse this item text; rows with cov_year = 2021 answered the same ENCIG 2021 question 5.4, which reads ‘Durante 2021’ and uses ungendered wording (e.g. ‘muchos pacientes o enfermos’).
mexico_2023_quality_healthservice	Fuente: INEGI, Encuesta Nacional de Calidad e Impacto Gubernamental (ENCIG) 2023 y ENCIG 2021 – IRW transcribed question 5.6/5.6a (health service of the State government or INSABI, formerly Seguro Popular) from INEGI’s 2023 questionnaire, stripped interviewer directions, shifted the response codes (Sí = 0, No = 1) and added an English translation of its own, and INEGI does not endorse this item text; rows with cov_year = 2021 answered the same ENCIG 2021 question 5.6, which reads ‘Durante 2021’ and uses ungendered wording (e.g. ‘muchos pacientes o enfermos’).
mexico_2023_quality_highways	Fuente: INEGI, Encuesta Nacional de Calidad e Impacto Gubernamental (ENCIG) 2023 y ENCIG 2021 – wording transcribed from INEGI’s 2023 questionnaire (question 5.13/5.13a), interviewer directions stripped, English translated by IRW, response codes shifted (Si = 0, No = 1), and no INEGI endorsement is implied; rows with cov_year = 2021 answered ENCIG 2021 question 5.11/5.11a, the same toll-highway items asked about 2021 with minor wording differences (‘etc?’ for ‘etcetera?’, ‘cuenta’ for ‘cuentan’, ungendered satisfaction labels).
mexico_2023_quality_homelightning	Fuente: INEGI, Encuesta Nacional de Calidad e Impacto Gubernamental (ENCIG) 2023 y ENCIG 2021 – IRW transcribed question 5.8/5.8a (the electricity service received at home) from INEGI’s 2023 questionnaire, stripped interviewer directions, shifted the response codes (Sí = 0, No = 1) and added an English translation of its own, and INEGI does not endorse this item text; rows with cov_year = 2021 answered the same question, numbered 5.7/5.7a in ENCIG 2021, which reads ‘Durante 2021’ and uses ungendered satisfaction labels.
mexico_2023_quality_lightning	The item text here is the administered Spanish, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. The table name misspells lighting, and is left as the published table spells it.
mexico_2023_quality_low	Fuente: INEGI, Encuesta Nacional de Calidad e Impacto Gubernamental (ENCIG) 2023 y ENCIG 2021 – IRW transcribed question 5.1 (whether the respondent used each on-demand public service during the year) from INEGI’s 2023 questionnaire, stripped interviewer directions, shifted the response codes (Sí = 0, No = 1) and added an English translation of its own, and INEGI does not endorse this item text; rows with cov_year = 2021 answered ENCIG 2021’s 5.1, which reads ‘durante 2021’, uses ungendered wording and did not ask items p5_1_06 (IMSS-Bienestar) or p5_1_10 (Cablebús/Mexicable).
mexico_2023_quality_parks	Fuente: INEGI, Encuesta Nacional de Calidad e Impacto Gubernamental (ENCIG) 2023 y ENCIG 2021 – IRW transcribed question 4.4/4.4a (parks and gardens) from INEGI’s 2023 questionnaire, stripped interviewer directions, shifted the response codes (Sí = 0, No = 1) and added an English translation of its own, and INEGI does not endorse this item text; rows with cov_year = 2021 answered the same question 4.4/4.4a, worded identically apart from ungendered satisfaction labels (Muy satisfecho … Muy insatisfecho).
mexico_2023_quality_police	Fuente: INEGI, Encuesta Nacional de Calidad e Impacto Gubernamental (ENCIG) 2023 y ENCIG 2021 – IRW transcribed question 4.6/4.6a (the police of the respondent’s city) from INEGI’s 2023 questionnaire, stripped interviewer directions, shifted the response codes (Sí = 0, No = 1) and added an English translation of its own, and INEGI does not endorse this item text; rows with cov_year = 2021 answered the same ENCIG 2021 question 4.6/4.6a, which uses ungendered wording (‘seguros’, ‘Muy satisfecho’).
mexico_2023_quality_problems	The item text here is the administered Spanish, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. Item texts are response categories of a three most important problems question, so each is a noun phrase rather than a sentence.
mexico_2023_quality_roads	Fuente: INEGI, Encuesta Nacional de Calidad e Impacto Gubernamental (ENCIG) 2023 y ENCIG 2021 – IRW transcribed question 4.7/4.7a (streets and avenues of the city) from INEGI’s 2023 questionnaire, stripped interviewer directions, shifted the response codes (Sí = 0, No = 1) and added an English translation of its own, and INEGI does not endorse this item text; rows with cov_year = 2021 answered the same question 4.7/4.7a, worded identically apart from ungendered satisfaction labels (Muy satisfecho … Muy insatisfecho).
mexico_2023_quality_schooling	Fuente: INEGI, Encuesta Nacional de Calidad e Impacto Gubernamental (ENCIG) 2023 y ENCIG 2021 – IRW transcribed question 5.2/5.2a (public primary, secondary and upper-secondary education) from INEGI’s 2023 questionnaire, stripped interviewer directions, shifted the response codes (Sí = 0, No = 1) and added an English translation of its own, and INEGI does not endorse this item text; rows with cov_year = 2021 answered the same question 5.2/5.2a, which reads ‘Durante 2021’ and uses ungendered wording (‘padres de familia’, ‘maestros’, ‘el alumno’, ‘los egresados’; Muy satisfecho … Muy insatisfecho).
mexico_2023_quality_streets	Fuente: INEGI, Encuesta Nacional de Calidad e Impacto Gubernamental (ENCIG) 2023 y ENCIG 2021 – IRW transcribed question 4.8/4.8a (toll-free highways and roads, ‘carreteras y caminos sin cuota’) from INEGI’s 2023 questionnaire, stripped interviewer directions, shifted the response codes (Sí = 0, No = 1) and added an English translation of its own, and INEGI does not endorse this item text; despite the table name, these items ask about the state’s toll-free highways and roads, not city streets (question 4.7, in mexico_2023_quality_roads); rows with cov_year = 2021 answered the same question 4.8/4.8a, worded identically apart from item p4_8_1 ending ‘etc?’ rather than ‘etcétera?’ and ungendered satisfaction labels (Muy satisfecho … Muy insatisfecho).
mexico_2023_quality_trains	Fuente: INEGI, Encuesta Nacional de Calidad e Impacto Gubernamental (ENCIG) 2023 y ENCIG 2021 – wording transcribed from INEGI’s 2023 questionnaire (question 5.12/5.12a, metro or light rail), interviewer directions stripped, English translated by IRW, response codes shifted (Si = 0, No = 1), and no INEGI endorsement is implied; rows with cov_year = 2021 answered ENCIG 2021 question 5.10/5.10a, the same metro/light-rail items asked about 2021 of the respondent’s city (‘(CIUDAD)’ rather than ‘(ESTADO)’), with option 2 worded ‘transcurre con poco tiempo entre un tren y otro?’ and ungendered satisfaction labels.
mexico_2023_quality_transportationstations	Fuente: INEGI, Encuesta Nacional de Calidad e Impacto Gubernamental (ENCIG) 2023 y ENCIG 2021 – IRW transcribed question 5.10/5.10a (articulated public transport with stations and an exclusive lane) from INEGI’s 2023 questionnaire, stripped interviewer directions, shifted the response codes (Sí = 0, No = 1) and added an English translation of its own, and INEGI does not endorse this item text; rows with cov_year = 2021 answered the same question numbered 5.9/5.9a in ENCIG 2021, which reads ‘Durante 2021’ and uses ungendered wording (‘pasajeros’, ‘cómodo’, ‘conductores’, ‘del operador hacia el usuario’; Muy satisfecho … Muy insatisfecho) and ‘transcurre con poco tiempo’ in item p5_10_3.
mexico_2023_quality_trash	Fuente: INEGI, Encuesta Nacional de Calidad e Impacto Gubernamental (ENCIG) 2023 y ENCIG 2021 – IRW transcribed question 4.5/4.5a (garbage collection) from INEGI’s 2023 questionnaire, stripped interviewer directions, shifted the response codes (Sí = 0, No = 1) and added an English translation of its own, and INEGI does not endorse this item text; rows with cov_year = 2021 answered the same question 4.5/4.5a, worded identically apart from ‘los empleados de limpia’ in item p4_5_2 and ungendered satisfaction labels (Muy satisfecho … Muy insatisfecho).
mexico_2023_quality_university	Fuente: INEGI, Encuesta Nacional de Calidad e Impacto Gubernamental (ENCIG) 2023 y ENCIG 2021 – IRW transcribed question 5.3/5.3a (public education in universities or technological universities) from INEGI’s 2023 questionnaire, stripped interviewer directions, shifted the response codes (Sí = 0, No = 1) and added an English translation of its own, and INEGI does not endorse this item text; rows with cov_year = 2021 answered the same question 5.3/5.3a, which reads ‘Durante 2021’ and uses ungendered wording (‘maestros’, ‘el alumno’, ‘pocos estudiantes’, ‘los egresados sean aptos’; Muy satisfecho … Muy insatisfecho).
mexico_2023_quality_water	The item text here is the administered Spanish, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors.
mexico_2023_quality_wellbeing	Fuente: INEGI, Encuesta Nacional de Calidad e Impacto Gubernamental (ENCIG) 2023 y ENCIG 2021 – IRW transcribed question 5.5/5.5a (health service in the ISSSTE) from INEGI’s 2023 questionnaire, stripped interviewer directions, shifted the response codes (Sí = 0, No = 1) and added an English translation of its own, and INEGI does not endorse this item text; despite the table name, the provider asked about is the ISSSTE, not IMSS-Bienestar or INSABI; rows with cov_year = 2021 answered the same ENCIG 2021 question 5.5, which reads ‘Durante 2021’ and uses ungendered wording (e.g. ‘muchos pacientes o enfermos’).
mexico_2023_quality_wellbeingservice	Fuente: INEGI, Encuesta Nacional de Calidad e Impacto Gubernamental (ENCIG) 2023 – IRW transcribed question 5.7/5.7a (health service of IMSS-Bienestar, formerly IMSS-Prospera) from INEGI’s 2023 questionnaire, stripped interviewer directions, shifted the response codes (Sí = 0, No = 1) and added an English translation of its own, and INEGI does not endorse this item text; the question was asked only in ENCIG 2023, of the roughly 430 respondents who reported using IMSS-Bienestar services.
mexico_2024_q1safety_conflicts	Fuente: INEGI, Encuesta Nacional de Seguridad Pública Urbana (ENSU) 2024, primer trimestre – IRW transcribed Section II (questions 2.1-2.4) from INEGI’s questionnaire, stripped interviewer directions, used INEGI’s data-catalogue labels for the 0/1 checklist codes, and added an English translation of its own; INEGI does not endorse this item text.
mexico_2024_q1safety_govperformance	Fuente: INEGI, Encuesta Nacional de Seguridad Pública Urbana (ENSU) 2024, primer trimestre – IRW transcribed Section III (questions 3.1-3.2a) from INEGI’s questionnaire, stripped interviewer directions, used INEGI’s data-catalogue labels for the 0/1 checklist codes, and added an English translation of its own; INEGI does not endorse this item text.
mexico_2024_q1safety_govtrust	Fuente: INEGI, Encuesta Nacional de Seguridad Pública Urbana (ENSU) 2024, primer trimestre – IRW transcribed Section IV (question 4.1) from INEGI’s questionnaire, stripped interviewer directions, left the unlabelled scale points 1-9 without option text, and added an English translation of its own; INEGI does not endorse this item text.
mexico_2024_q1safety_perception	Fuente: INEGI, Encuesta Nacional de Seguridad Pública Urbana (ENSU) 2024, primer trimestre – IRW transcribed Section I (questions 1.1-1.9) from INEGI’s questionnaire, stripped interviewer directions, took the 0/1 labels for the question 1.6 information-source indicators (including its ‘No sabe/ no responde’ and ‘Otra’ indicators, which are shipped as items) from INEGI’s data catalogue, and added an English translation of its own; INEGI does not endorse this item text.
mexico_2024_q2safety_conflicts	Fuente: INEGI, Encuesta Nacional de Seguridad Pública Urbana (ENSU) 2024, segundo trimestre – IRW transcribed Section II (questions 2.1-2.4) from INEGI’s questionnaire, stripped interviewer directions, used INEGI’s data-catalogue labels for the 0/1 checklist codes, and added an English translation of its own; INEGI does not endorse this item text.
mexico_2024_q2safety_govperformance	Fuente: INEGI, Encuesta Nacional de Seguridad Pública Urbana (ENSU) 2024, segundo trimestre – IRW transcribed Section III (questions 3.1-3.6) from INEGI’s questionnaire, stripped interviewer directions, used INEGI’s data-catalogue labels for the 0/1 checklist codes, and added an English translation of its own; INEGI does not endorse this item text.
mexico_2024_q2safety_harassment	Fuente: INEGI, Encuesta Nacional de Seguridad Pública Urbana (ENSU) 2024, segundo trimestre – IRW transcribed Section IV (question 4.1) from INEGI’s questionnaire, stripped interviewer directions, and added an English translation of its own; INEGI does not endorse this item text.
mexico_2024_q2safety_perception	Fuente: INEGI, Encuesta Nacional de Seguridad Pública Urbana (ENSU) 2024, segundo trimestre – IRW transcribed Section I (questions 1.1-1.9) from INEGI’s questionnaire, stripped interviewer directions (including the 1.6 notes limiting vehicle thefts to household-owned vehicles and incidents 4-6 to household members aged 18+), and added an English translation of its own; INEGI does not endorse this item text.
mexico_2024_q3safety_conflicts	Fuente: INEGI, Encuesta Nacional de Seguridad Pública Urbana (ENSU) 2024, tercer trimestre – IRW transcribed Section II (questions 2.1-2.4) from INEGI’s questionnaire, stripped interviewer directions, used INEGI’s data-catalogue labels for the 0/1 checklist codes, and added an English translation of its own; INEGI does not endorse this item text.
mexico_2024_q3safety_govperformance	Fuente: INEGI, Encuesta Nacional de Seguridad Pública Urbana (ENSU) 2024, tercer trimestre – IRW transcribed Section III (questions 3.1-3.2a) from INEGI’s questionnaire, stripped interviewer directions, used INEGI’s data-catalogue labels for the 0/1 checklist codes, and added an English translation of its own; INEGI does not endorse this item text.
mexico_2024_q3safety_govtrust	Fuente: INEGI, Encuesta Nacional de Seguridad Pública Urbana (ENSU) 2024, tercer trimestre – IRW transcribed Section V (question 5.1) from INEGI’s questionnaire, stripped interviewer directions, and added an English translation of its own; INEGI does not endorse this item text. Code 1 is the most trust (Mucha confianza) and 4 the least, the reverse of the 0-10 trust item in the 2024 first-quarter table.
mexico_2024_q3safety_perception	Fuente: INEGI, Encuesta Nacional de Seguridad Pública Urbana (ENSU) 2024, tercer trimestre – IRW transcribed Section I (questions 1.1-1.9) from INEGI’s questionnaire, stripped interviewer directions, took the 0/1 labels for the question 1.6 information-source indicators (including its ‘No sabe / no responde’ and ‘Otra’ indicators, which are shipped as items) from INEGI’s data catalogue, and added an English translation of its own; INEGI does not endorse this item text.
mexico_2024_q3safety_violence	Fuente: INEGI, Encuesta Nacional de Seguridad Pública Urbana (ENSU) 2024, tercer trimestre – IRW transcribed Section IV question 4.3 from INEGI’s questionnaire, stripped interviewer directions, and added an English translation of its own; the items ask about situations the respondent experienced with members of their own household, not violent crime in general; INEGI does not endorse this item text.
mexico_2024_q4safety_conflicts	Fuente: INEGI, Encuesta Nacional de Seguridad Pública Urbana (ENSU) 2024, cuarto trimestre – IRW transcribed Section II (questions 2.1-2.4) from INEGI’s questionnaire, stripped interviewer directions, used INEGI’s data-catalogue labels for the 0/1 checklist codes, and added an English translation of its own; INEGI does not endorse this item text.
mexico_2024_q4safety_govperformance	Fuente: INEGI, Encuesta Nacional de Seguridad Pública Urbana (ENSU) 2024, cuarto trimestre – IRW transcribed Section III (questions 3.1-3.6) from INEGI’s questionnaire, stripped interviewer directions, used INEGI’s data-catalogue labels for the 0/1 checklist codes, and added an English translation of its own; INEGI does not endorse this item text.
mexico_2024_q4safety_harassment	Fuente: INEGI, Encuesta Nacional de Seguridad Pública Urbana (ENSU) 2024, cuarto trimestre – IRW transcribed Section IV (question 4.1) from INEGI’s questionnaire, stripped interviewer directions, and added an English translation of its own; INEGI does not endorse this item text.
mexico_2024_q4safety_perception	Fuente: INEGI, Encuesta Nacional de Seguridad Pública Urbana (ENSU) 2024, cuarto trimestre – IRW transcribed Section I (questions 1.1-1.9) from INEGI’s questionnaire, stripped interviewer directions (including the 1.6 notes limiting vehicle thefts to household-owned vehicles and incidents 4-8 to household members aged 18+), and added an English translation of its own; INEGI does not endorse this item text.
mexico_2025_q1safety_conflicts	Fuente: INEGI, Encuesta Nacional de Seguridad Pública Urbana (ENSU) 2025, primer trimestre – IRW transcribed Section II (questions 2.1-2.4) from INEGI’s questionnaire, stripped interviewer directions, used INEGI’s data-catalogue labels for the 0/1 checklist codes, and added an English translation of its own; INEGI does not endorse this item text.
mexico_2025_q1safety_govperformance	Fuente: INEGI, Encuesta Nacional de Seguridad Pública Urbana (ENSU) 2025, primer trimestre – IRW transcribed Section III (questions 3.1-3.2a) from INEGI’s questionnaire, stripped interviewer directions, used INEGI’s data-catalogue labels for the 0/1 checklist codes, and added an English translation of its own; INEGI does not endorse this item text.
mexico_2025_q1safety_govtrust	Fuente: INEGI, Encuesta Nacional de Seguridad Pública Urbana (ENSU) 2025, primer trimestre – IRW transcribed Section IV (question 4.1 and its read-aloud introduction) from INEGI’s questionnaire, stripped interviewer directions, and added an English translation of its own; INEGI does not endorse this item text. Code 1 is the most trust (Mucha confianza) and 4 the least; the items share codes bp4_1_1..3 with the 2024 first-quarter table but not its question or its 0-10 scale.
mexico_2025_q1safety_perception	Fuente: INEGI, Encuesta Nacional de Seguridad Pública Urbana (ENSU) 2025, primer trimestre – IRW transcribed Section I (questions 1.1-1.9) from INEGI’s questionnaire, stripped interviewer directions, took the 0/1 labels for the question 1.6 information-source indicators (including its ‘No sabe/ no responde’ and ‘Otra’ indicators, which are shipped as items) from INEGI’s data catalogue, and added an English translation of its own; INEGI does not endorse this item text.
mexico_2025_q2safety_conflicts	Fuente: INEGI, Encuesta Nacional de Seguridad Pública Urbana (ENSU) 2025, segundo trimestre – IRW transcribed Section II (questions 2.1-2.4) from INEGI’s questionnaire, stripped interviewer directions, used INEGI’s data-catalogue labels for the 0/1 checklist codes, and added an English translation of its own; INEGI does not endorse this item text.
mexico_2025_q2safety_govperformance	Fuente: INEGI, Encuesta Nacional de Seguridad Pública Urbana (ENSU) 2025, segundo trimestre – IRW transcribed Section III (questions 3.1-3.6) from INEGI’s questionnaire, stripped interviewer directions, used INEGI’s data-catalogue labels for the 0/1 checklist codes, and added an English translation of its own; INEGI does not endorse this item text.
mexico_2025_q2safety_harassment	Fuente: INEGI, Encuesta Nacional de Seguridad Pública Urbana (ENSU) 2025, segundo trimestre – IRW transcribed Section IV (question 4.1) from INEGI’s questionnaire, stripped interviewer directions, and added an English translation of its own; INEGI does not endorse this item text.
mexico_2025_q2safety_perception	Fuente: INEGI, Encuesta Nacional de Seguridad Pública Urbana (ENSU) 2025, segundo trimestre – IRW transcribed Section I (questions 1.1-1.9) from INEGI’s questionnaire, stripped interviewer directions (including the 1.6 notes limiting vehicle thefts to household-owned vehicles and incidents 4-8 to household members aged 18+), and added an English translation of its own; INEGI does not endorse this item text.
mexico_2025_q3safety_conflicts	Fuente: INEGI, Encuesta Nacional de Seguridad Pública Urbana (ENSU) 2025, tercer trimestre – IRW transcribed Section II (questions 2.1-2.4) from INEGI’s questionnaire, stripped interviewer directions, used INEGI’s data-catalogue labels for the 0/1 checklist codes, and added an English translation of its own; INEGI does not endorse this item text.
mexico_2025_q3safety_govperformance	Fuente: INEGI, Encuesta Nacional de Seguridad Pública Urbana (ENSU) 2025, tercer trimestre – IRW transcribed Section III (questions 3.1-3.2a) from INEGI’s questionnaire, stripped interviewer directions, used INEGI’s data-catalogue labels for the 0/1 checklist codes, and added an English translation of its own; INEGI does not endorse this item text.
mexico_2025_q3safety_govtrust	Fuente: INEGI, Encuesta Nacional de Seguridad Pública Urbana (ENSU) 2025, tercer trimestre – IRW transcribed Section V (question 5.1 and its read-aloud introduction) from INEGI’s questionnaire, stripped interviewer directions, and added an English translation of its own; INEGI does not endorse this item text. Code 1 is the most trust (Mucha confianza) and 4 the least; in this quarter the items are coded bp5_1_1..3, while the same question in the 2025 first-quarter table is bp4_1_1..3.
mexico_2025_q3safety_perception	Fuente: INEGI, Encuesta Nacional de Seguridad Pública Urbana (ENSU) 2025, tercer trimestre – IRW transcribed Section I (questions 1.1-1.9) from INEGI’s questionnaire, stripped interviewer directions, took the 0/1 labels for the question 1.6 information-source indicators (including its ‘No sabe / no responde’ and ‘Otra’ indicators, which are shipped as items) from INEGI’s data catalogue, and added an English translation of its own; INEGI does not endorse this item text. In this quarter question 1.2 asks about ‘la calle’ (bp1_2_03) and a defined ‘parque o centro recreativo’ (bp1_2_12), not the ‘calles que habitualmente usa’ and ‘parque recreativo o centro recreativo’ of the 2024 tables and the 2025 first-quarter table.
mexico_2025_q3safety_violence	Fuente: INEGI, Encuesta Nacional de Seguridad Pública Urbana (ENSU) 2025, tercer trimestre – IRW transcribed Section IV question 4.3 from INEGI’s questionnaire, stripped interviewer directions, and added an English translation of its own; the items ask about situations the respondent experienced with members of their own household, not violent crime in general; INEGI does not endorse this item text.
mexico_2025_q4safety_conflicts	Fuente: INEGI, Encuesta Nacional de Seguridad Pública Urbana (ENSU) 2025, cuarto trimestre – IRW transcribed Section II (questions 2.1-2.4) from INEGI’s questionnaire, stripped interviewer directions, used INEGI’s data-catalogue labels for the 0/1 checklist codes, and added an English translation of its own; INEGI does not endorse this item text.
mexico_2025_q4safety_govperformance	Fuente: INEGI, Encuesta Nacional de Seguridad Pública Urbana (ENSU) 2025, cuarto trimestre – IRW transcribed Section III (questions 3.1-3.6) from INEGI’s questionnaire, stripped interviewer directions, used INEGI’s data-catalogue labels for the 0/1 checklist codes, and added an English translation of its own; INEGI does not endorse this item text.
mexico_2025_q4safety_harassment	Fuente: INEGI, Encuesta Nacional de Seguridad Pública Urbana (ENSU) 2025, cuarto trimestre – IRW transcribed Section IV (question 4.1) from INEGI’s questionnaire, stripped interviewer directions, and added an English translation of its own; INEGI does not endorse this item text.
mexico_2025_q4safety_perception	Fuente: INEGI, Encuesta Nacional de Seguridad Pública Urbana (ENSU) 2025, cuarto trimestre – IRW transcribed Section I (questions 1.1-1.9) from INEGI’s questionnaire, stripped interviewer directions (including the 1.6 notes limiting vehicle thefts to household-owned vehicles and incidents 4-8 to household members aged 18+), and added an English translation of its own; INEGI does not endorse this item text. In this quarter the extortion incidence (bp1_6_6) reads ‘Presiones, engaños o advertencias’ where the 2024 fourth-quarter and 2025 second-quarter tables read ‘Amenazas, presiones o engaños’, and question 1.2 asks about ‘la calle’ (bp1_2_03) and a defined ‘parque o centro recreativo’ (bp1_2_12), unlike the 2024 tables.
mft_validationandreplication	Item text is the MFQ-2 as published by its originators (Atari et al.); the study’s own paper could not be checked, and for MFQ_Prop5 the originators also circulated an alternative wording (‘In a fair society, I want people who work harder than others to end up richer than others’), so which version respondents saw is not confirmed.
mgkt	The correct answers were not published with this dataset: the codebook gives each question’s text and its ten alternatives but never states which five are correct. The key shipped in correct_response was derived by this project from the response data, not taken from the source. Each MGKT question showed 10 alternatives of which exactly 5 were correct, and respondents checked as many as they knew; the IRW resp is therefore not a chosen option but a per-item score, the number of correct alternatives checked minus the number of incorrect ones (-5 to 5), which is why option_text is blank. correct_response lists the five correct alternatives for each question as text, not as resp values. Note that the administered instructions shipped here state a -1.25 penalty per wrong answer (the feedback scoring on the test page), while the score stored in the IRW table applies a penalty of exactly 1 per wrong answer; this was confirmed against the released raw data for all 32 items. The five incorrect decoy alternatives per question are not shipped here but are listed in the source codebook.
mhscdc_fried_2020_ema	This EMA table carries TWO different 1-5 response ladders and nothing in the resp values distinguishes them: Q1-Q10 are momentary affect items scored 1 ‘Not at all’ to 5 ‘Extremely’, while Q11-Q18 are duration items scored 1 ‘0 min’, 2 ‘1 - 15 min’, 3 ‘15 - 60 min’, 4 ‘1 - 2 hours’, 5 ‘> 2 hours’. Do not read one common scale across the table. Item wording is transcribed from the deposit’s administered survey form (Measures_EMA.pdf), which differs in several places from the paraphrased Item column of Codebook_EMA.xlsx.
miedema_2023_ecs40	The interview was conducted in Bangla, which the deposit does not publish in full; the item text is the study’s English rendering from its data-file labels. Those labels are cut at 80 characters, so 17 item texts (ecoscale_8, 10, 12, 13, 16, 18, 19, 27, 29, 31, 33, 43, 45, 46, 47, 49, 50) end mid-sentence; their full wording is only in a closed-access paper. Another 19 were completed by IRW from the paper’s Table 1. The Yes/No options are the administered Bangla, with English alongside.
milavic_2019_psisysf	Item wording for this table is English while the instrument was administered in Croatian: the study’s data deposit and its single supplementary file carry no Croatian text, so the English printed in the paper’s Table 1 is shipped in its place, and the anchors in option_text (1 = strongly disagree, 5 = strongly agree) are taken from that table’s footnote even though the same paper’s Methods section instead describes the scale as running from 1 = almost never to 5 = almost always
milson_2026_self_esteem	Item text is the canonical English Rosenberg Self-Esteem Scale; the study’s deposit carries no item wording or labels, so the assignment of Rosenburg_esteem_1..10 to items follows the commonly circulated RSES numbering (not the different order printed on the University of Maryland sheet), which the data support by keying structure and marker items but do not pin item-by-item. Responses are stored already reverse-scored (higher = higher self-esteem), so the anchors shown for items 2, 5, 6, 8 and 9 run from Strongly Agree (1) to Strongly Disagree (4).
mironshatz_2015_happiness_beliefs	Items state1 and state6 are truncated at 80 characters in the source data file and are shipped as those fragments (‘…have all we woul’, ‘…productive act’); the full wording is not published. No response-option labels were published for the 1-4 scale, so the direction of the scale is unknown.
mistry_2022_hardship	For mistry_2022_hardship the option_text values are the labels of the collapsed binary recode the deposit ships (‘No difficulty’/‘Some difficulty’), not the response options respondents were offered, which for the food, medicine and routine-care items were None/some/much/unable or very difficult; and the item_text for memory and difficulty_earning is cut off mid-phrase because the source Stata variable labels hit Stata’s 80-character cap
mobility	The item wording for this table is the English text documented in the ltm R package, not the Bengali in which the Bangladesh Fertility Survey 1989 was administered to rural Bangladeshi women; no Bengali original is published in the source, and the instructions field is the documentation’s description of what respondents were asked rather than a verbatim administered instruction.
modzelewska_2021_emo_neg	Item wording is the Polish emotion word respondents rated, with English alongside, but the instruction sentence and the two endpoint anchors are the study’s English translation because the Polish questionnaire text was not published.
modzelewska_2021_emo_pos	Item wording is the Polish emotion word respondents rated, with English alongside, but the instruction sentence and the two endpoint anchors are the study’s English translation because the Polish questionnaire text was not published.
modzelewska_2021_hbeliefs	Item wording is the Polish statement respondents rated, with the study’s English alongside, but the instruction text and the two endpoint anchors are the study’s English translation because the Polish questionnaire text was not published.
moe2025_erq	The ERQ was administered to Italian teachers, but the figshare deposit is a single SPSS file with no item wording in any language and no accompanying article or supplement could be located, so the item text shown here is Gross & John’s (2003) canonical English ERQ rather than the Italian sentences the teachers read. Item codes ERQ1_pre-ERQ10_pre are matched to the instrument’s own printed item numbers; the assignment of reappraisal versus suppression wording is confirmed by the deposit’s own Reappraisal composite, which is the mean of exactly ERQ1/3/5/7/8/10 (the unique such six-item subset of 210, next-best deviation 0.83), but nothing in the data distinguishes the near-parallel reappraisal items 1, 3, 7 and 10 from one another. Scale points 2, 3, 5 and 6 carry no option text because the ERQ form labels only 1, 4 and 7.
moe2025_scs	The scale was administered to Italian teachers, but the deposit is an SPSS file with no item wording in any language. The Italian in item_text is the published Italian Self-Compassion Scale from the instrument’s own site (self-compassion.org), and the English in the _translated columns is Neff’s original SCS-26, so neither comes from the study’s own materials. Codes are matched to the instrument’s printed item numbers.
mohammed_2021_job_satisfaction	Item text and response anchors are transcribed exactly as the study’s SPSS file stores them – all lowercase, no terminal punctuation, and one item reads “am proud to work at this hospital” with no leading “I”. The wording has deliberately not been tidied. The source study never reproduces these five items and never names the instrument: its identification as the Safety Attitudes Questionnaire job-satisfaction domain is inferred from the wording, not stated by the study, and the 1 = strongly disagree to 5 = strongly agree direction rests on the SPSS value labels alone, which nothing else in the study could corroborate
mohammed_2021_patient_safety_culture	The only item wording this study ever published is the variable labels inside its SPSS data file, and those labels are truncated: they stop at 70 characters (so some items lose their final words) and they are also cut at the start, so many begin with a word fragment such as ‘fied.’ or ‘.’. They are shipped exactly as the file stores them rather than repaired, because this administration is not the canonical AHRQ HSOPSC: 18 items are negated rewrites of AHRQ’s reverse-worded items, and the responses run in the labels’ direction (all 18 correlate +0.13 to +0.55 with the other items rather than negatively), so substituting the standard wording would have inverted them.
molino_2018_d5_scale	The table name signals one scale but the table pools two measures from section 5 of the questionnaire: d_5_1-d_5_5 are entrepreneurial intention (adapted from Linan & Chen, 2009) and d_5_6-d_5_8 are perceived support from family and friends. Item text is the administered Italian from the S2 Questionnaire, with the authors’ own English in the _translated columns. Only the endpoints of the 7-point scale are labelled, so option_text is blank for 2-6.
monier_2026_pot	Item text for this table is the study authors’ English rendering; the questions were administered in French, and no French original is published in the paper or its S1 Table, so the administered wording could not be shipped.
monier_2026_sti	Item text for this table is the English wording printed in the study’s own Table 6; the inventory was administered in French, and no full French original is published in the paper or its S1 Table, so the administered wording could not be shipped. The paper never states the 7-point response anchors, so option text is blank.
montana_blasco_2026_content_freq	As for montana_blasco_2026_digital_skills: IRW’s English translation of the Spanish codebook wording.
montana_blasco_2026_digital_skills	The English item and option text (*_translated) was written by IRW from the Spanish codebook wording; it is not the authors’ wording.
montana_blasco_2026_sm_impact	As for montana_blasco_2026_digital_skills: IRW’s English translation of the Spanish codebook wording.
montana_blasco_2026_tiktok_wellbeing	As for montana_blasco_2026_digital_skills: IRW’s English translation of the Spanish codebook wording.
moon_2023_korean_proficiency	Item text is the study’s English questionnaire; most respondents (e.g. 159 of 206 Vietnamese nationals) answered Korean, Vietnamese, Chinese or Filipino versions that were not published, and the paper’s text describes the scale as ‘1 = not at all to 4 = very good’ whereas the questionnaire and data file use five points from 1 = Never to 5 = Very good.
moon_2023_pregnancy_stress	The questionnaire was administered in Korean, Vietnamese, Chinese, Filipino and English (77% of respondents were Vietnamese nationals); the item text is the study’s English form, the only language version in the deposit, and the scale’s middle options 2 and 3 were unlabelled.
moon_2023_selfesteem	Item text is the study’s English questionnaire, but most respondents (77% were Vietnamese nationals) answered the authors’ unpublished Korean, Vietnamese, Chinese or Filipino translations; the five negatively worded items (se3, se5, se9, se10, selfesteem11) are stored already reverse-scored, so 4 means ‘not at all’, while se8 (‘I would like to have more self-respect’) is scored in its raw direction, unlike the Rosenberg original.
moon_2023_spousal_support	Item text is the study’s English questionnaire; most respondents (159 of 206 were Vietnamese nationals) answered the authors’ unpublished Korean, Vietnamese, Chinese or Filipino translations, and only the endpoints 1 (Very dissatisfied) and 6 (Very satisfied) are labelled.
morales_2021_erq	The ERQ was administered in Spanish (the Cabello et al., 2013 adaptation) to a mostly Chilean sample, but neither the PeerJ deposit nor the paper’s supplement contains any Spanish item wording – the deposit’s ERQ columns are bare codes ERQ_1..ERQ_10 – so the item text shown here is Gross and John’s (2003) canonical English ERQ rather than the sentences participants read. Item codes erq1-erq10 follow the instrument’s printed item numbers; the split into reappraisal (erq1, 3, 5, 7, 8, 10) and suppression (erq2, 4, 6, 9) wording is confirmed by the paper’s Table S.5, whose twelve published subscale statistics reproduce exactly under that key and under no other of the 210 possible six-item splits, but nothing in the data distinguishes the near-parallel reappraisal items 1, 3, 7 and 10 from one another. Scale points 2, 3, 5 and 6 carry no option text because the ERQ form labels only 1, 4 and 7.
morano_2019_self_efficacy	Each item is a row of four graded statements with pictograms, not a question, so item_text is blank and the four statements are in option_text (resp 1 = lowest efficacy, 4 = highest). The pictograms are not reproduced. The administered Italian is in the base fields, with the authors’ own English alongside.
motion	The items of motion are stimulus conditions in a random-dot motion task, not questions: each item code is block number and motion coherence (e.g. ‘1 6’ = block 1, 6% coherence), and the item_text IRW ships is a description of that stimulus rather than wording the child read. The only verbatim on-screen text is in the instructions field. resp is scored accuracy (1 = correct direction judgement), not the left/right key the child pressed.
mpsycho_avlancheprep	Item text is the terse stem the MPsychoR documentation gives for each column (info, discuss, gear, decision), not the full survey question skiers read; the Haegeli et al. (2012) paper was not available. Which option_text belongs to which resp value is inferred from the order the documentation lists the options in, since no numeric key is published: the data confirm the most-prepared and least-prepared ends for all four items, but the order of the middle categories is unverified.
mturkddm_by	A dot-numerosity task: an array of yellow and blue dots was shown and the participant chose which colour was more numerous. The item code gives the two counts, yellow first – so ‘10 15’ means ten yellow dots and fifteen blue. resp is whether the choice was correct. The total dot area was also varied in the original experiment but is not part of the item code, so trials differing only in area are pooled here.
mturkddm_lexical	A lexical decision task: 4,588 letter strings, 2,295 real words and 2,293 nonwords, each flashed briefly on screen. The item code is the string itself. Participants pressed the right arrow for a real word and the left arrow for a nonword, and resp records whether that judgement was correct, so each string is listed twice – once for the response that was correct for it and once for the response that was not. Items are grouped by the word-frequency pool they were drawn from.
much_tte_2025_concentrationtask	Nine concentration items. Each showed a grid of visually confusable letters and one target letter, and the participant counted how many times the target appeared. The description here names the target letter and the true number of occurrences, which is also given as the correct answer; the grid itself was a picture and is not reproduced. Responses are 0 or 1 for whether the count given was right. The items came in three blocks of three.
much_tte_2025_currentmotivation	Three statements about how much effort the participant intended to put into the test block they were about to start, answered on a sliding scale from ‘strongly disagree’ to ‘strongly agree’ with 101 internal points. The number was not shown to the participant while answering. Two further items from the same source scale were dropped from this version because they could not be sensibly rephrased in the future tense.
much_tte_2025_effort	Five statements about how much effort the participant put into the test block they had just finished, answered from 1 (strongly disagree) to 4 (strongly agree). Only the two ends of the scale were labelled. The same five items were asked after each block; which block a given answer refers to is recorded in the response data rather than in the item.
much_tte_2025_matrixreasoning	The speed/accuracy block instruction was counterbalanced across participants (Group 1 saw the non-speeded framing first, Group 2 the speeded), so the instructions field reproduces both verbatim rather than the one wording any single respondent read; item_text is blank because the OMIB matrix stimuli are figures, not text.
muslih_2024_rses	The item text shipped for this table is the English wording the paper prints in its Table 2; the scale was administered to participants in Indonesian, and no Indonesian wording is published in the paper or its S1 Dataset.
najari_2024_bpqsf_awareness	This table is named for the BPQ-SF’s Body Awareness subscale but actually contains the whole 46-item BPQ-SF: q1-q26 are Body Awareness, q27-q40 Supradiaphragmatic Reactivity and q41-q46 Subdiaphragmatic Reactivity, per Table 2 of the source paper.
nam_2024_function	The survey was administered online in Korean and neither the article nor its S1 Data supplement publishes any Korean wording, so the item text shown here is the Washington Group’s own canonical English WG-SS wording; the assignment of FL1-FL6 to the six domains follows the canonical WG-SS order that the paper’s Methods reproduces, and the data confirm the mobility and self-care items (FL3, FL5) and the seeing/hearing pair (FL1, FL2) but cannot tell seeing from hearing.
namprb_siwiak_2024_aot	The item text here is the administered Polish, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. The source carries PDF line-break hyphens inside words (naszy-mi, przeko-nan, na-wet). These are kept verbatim in the administered text and rendered whole in the translation.
namprb_siwiak_2024_bsr10	The item text here is the administered Polish, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. Pennycook’s English originals exist, but these are Polish renderings of randomly generated pseudo-profound statements, and the construct is the surface plausibility of the wording itself – so substituting the original would change what the item measures.
namprb_siwiak_2024_kon2066	The item text here is the administered Polish, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors.
namprb_siwiak_2024_kop20	The item text here is the administered Polish, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. The instructions and all four option labels have had their Polish diacritics stripped by an encoding fault in the source (W naszym spolecze stwie, Zgadzam si). The damage is left exactly as published, because it is what the table currently serves; the translation renders the intended sense.
namprb_siwiak_2024_ssub	IRW table doesn’t match the number of items provided in the paper
NAMPRB_Siwiak_2024_SSUB	The table carries 25 item codes for a 20-item scale. The five religious items (1, 5, 14, 15, 19) appear as SSUB_TRB_n at wave 0 and as SSUB_NAB_n at wave 1, because the retest columns were renamed in bulk; each pair is the same question and carries the same text here. The items are the administered Polish with the authors’ English alongside, but the anchors (Strongly disagree to Strongly agree) are the English published in the study’s preprint, not the Polish respondents saw. The ten reverse-scored items (3, 5, 7, 8, 10, 13, 15, 16, 17, 18) are stored unreversed.
narcissistic_personality_inventory	This is the forced-choice NPI-40, so item_text is blank for all 40 items by design and both statements of each pair are carried as option_text at resp 1 and 2. The 80 statements are the 2014 deposit’s own codebook wording. The instructions line, however, was transcribed from the live openpsychometrics NPI form as it stood in 2026, because the deposit’s codebook only describes how the test was administered rather than quoting the instruction respondents read; the 80 item statements on that live form are byte-identical to the 2014 codebook, so it is the same instrument, but the instruction wording itself is current-page evidence for a 2014 administration.
nas_rogoza_2024_study2_ngs	Thirteen adjectives from the Narcissistic Grandiosity Scale, answered from 1 (not at all) to 7 (extremely) for how well each word describes you in general. They were administered as part of a longer list of 42 narcissism adjectives, of which these are the grandiosity block. The words are shown in English; the study was run in Polish and the Polish wording is in the deposit but not reproduced here. Which adjective is item 1 versus item 2 follows the order the deposit’s codebook prints them in.
nas_rogoza_2024_study5_nas	Study 5 was a Polish daily diary, so the item text is the administered Polish with English alongside; the four English adjectives and the 0-100 slider anchors are the authors’ own from their OSF NAS.xlsx, but the daily instruction was never published in English and its English here was produced by IRW. The deposited .sav has no labels, so the adjective assigned to NAS1-NAS4 follows the order of the study’s daily codebook (which differs from the 16-item form’s order). The data confirm NAS2 (Lekceważący / Depreciating) and NAS3 (Wyzyskujący / Exploitative) but cannot tell NAS1 from NAS4 (Znieważający / Abusive vs Bezczelny / Nasty), whose means differ by 0.5 points on the 0-100 slider.
nas_rogoza_2024_study5_ngs	Study 5 was a Polish daily diary, so the item text is the administered Polish with English alongside; the English adjectives and the 0/100 anchors are the authors’ own from the paper, but the daily instruction was never published in English and its English here was produced by IRW. The deposited .sav has no labels, so the assignment of NGS1-NGS4 to Brilliant, Glorious, Powerful and Prestigious follows the order in the study’s daily codebook and the paper. The data confirm the split between the two higher-endorsed adjectives (NGS1, NGS2) and the two lower ones (NGS3, NGS4), but cannot distinguish Brilliant from Glorious or Powerful from Prestigious.
nas_rogoza_2024_study5_nvs	Study 5 was a Polish daily diary, so the item text is the administered Polish with English alongside; the English adjectives and anchors are the authors’ own (from the article and the deposit’s NAS.xlsx), but the daily instruction was never published in English and its English here was produced by IRW. The deposited .sav has no labels, so the assignment of NVS1-NVS4 to Ignored, Resentful, Misunderstood and Underappreciated follows the order three of the study’s documents share; the data confirm NVS3 and NVS4 as the more strongly endorsed pair but cannot separate Ignored from Resentful or Misunderstood from Underappreciated. NVS1’s Polish ‘Zleckewazony’ is reproduced as both state-protocol codebooks spell it; it is a misspelling of ‘Zlekcewazony’.
nature_relatedness	Only items 1-6 of this table are Nature Relatedness Scale (NR-6) items. Items 7-16 are the Ten Item Personality Inventory, and items 17-32 are a 16-word vocabulary check-list administered in the same session (three of its words, ‘cuivocal’, ‘florted’ and ‘verdid’, are deliberate non-words). The three blocks can be told apart by their response ranges (1-5, 1-7 and 0-1).
neurodegenerative_huizinga_2019_dass	The study does not say which Dutch DASS-21 translation it administered, and two are distributed that differ in items 1-3; IRW ships the 2010 revised de Beurs version, chosen because DASS.1 behaves as a stress item (corrected r .618 stress vs .448 anxiety, where the 2001 form’s item 1 loads equally on both) and DASS.2 behaves as the dry-mouth item (weakest item in the scale, r .329, yet the most endorsed anxiety item at mean .502) rather than the 2001 form’s sweating item; the instructions are the paper form’s and mention circling a number, though this battery was administered online
neurodegenerative_huizinga_2019_vfq	Item text is the NEI VFQ-25’s official ENGLISH wording; the study administered the questionnaire in Dutch and neither the paper’s appendices nor the DataverseNL deposit publish the Dutch form, so this is a same-instrument substitute rather than what respondents read. Note also that resp is the RAND-converted 0-100 item score rescaled (higher = better vision- related functioning), not the response number printed on the questionnaire, so for items 1-14, 15c, 16 and 16a the option order is the reverse of the printed one.
nguyen_2026_autonomy_academic_motivation	Item text is the English questionnaire the author deposited with the data; the survey of Hanoi lower-secondary students was very likely administered in Vietnamese, and that version is not in the deposit. Separately, the constructs in this deposit fall into two blocks that correlate negatively with each other (academic pressure, academic motivation and family factors against school autonomy support, voice, choice and ownership), and some item pairs correlate in the direction opposite to their wording. That pattern would fit one block having been stored reverse to the questionnaire’s 1 = Strongly disagree to 5 = Strongly agree, but there is no article to check it against, so the anchors ship as the questionnaire states them.
nguyen_2026_autonomy_academic_pressure	Item text is the English questionnaire the author deposited with the data; the survey of Hanoi lower-secondary students was very likely administered in Vietnamese, and that version is not in the deposit. Separately, the constructs in this deposit fall into two blocks that correlate negatively with each other (academic pressure, academic motivation and family factors against school autonomy support, voice, choice and ownership), and some item pairs correlate in the direction opposite to their wording. That pattern would fit one block having been stored reverse to the questionnaire’s 1 = Strongly disagree to 5 = Strongly agree, but there is no article to check it against, so the anchors ship as the questionnaire states them.
nguyen_2026_autonomy_family_factors	Item text is the English questionnaire the author deposited with the data; the survey of Hanoi lower-secondary students was very likely administered in Vietnamese, and that version is not in the deposit. Separately, the constructs in this deposit fall into two blocks that correlate negatively with each other (academic pressure, academic motivation and family factors against school autonomy support, voice, choice and ownership), and some item pairs correlate in the direction opposite to their wording. That pattern would fit one block having been stored reverse to the questionnaire’s 1 = Strongly disagree to 5 = Strongly agree, but there is no article to check it against, so the anchors ship as the questionnaire states them.
nguyen_2026_autonomy_school_autonomy_support	Item text is the English questionnaire the author deposited with the data; the survey of Hanoi lower-secondary students was very likely administered in Vietnamese, and that version is not in the deposit.
nguyen_2026_autonomy_student_choice	Item text is the English questionnaire the author deposited with the data; the survey of Hanoi lower-secondary students was very likely administered in Vietnamese, and that version is not in the deposit. Separately, the constructs in this deposit fall into two blocks that correlate negatively with each other (academic pressure, academic motivation and family factors against school autonomy support, voice, choice and ownership), and some item pairs correlate in the direction opposite to their wording. That pattern would fit one block having been stored reverse to the questionnaire’s 1 = Strongly disagree to 5 = Strongly agree, but there is no article to check it against, so the anchors ship as the questionnaire states them.
nguyen_2026_autonomy_student_ownership	Item text is the English questionnaire the author deposited with the data; the survey of Hanoi lower-secondary students was very likely administered in Vietnamese, and that version is not in the deposit. Separately, the constructs in this deposit fall into two blocks that correlate negatively with each other (academic pressure, academic motivation and family factors against school autonomy support, voice, choice and ownership), and some item pairs correlate in the direction opposite to their wording. That pattern would fit one block having been stored reverse to the questionnaire’s 1 = Strongly disagree to 5 = Strongly agree, but there is no article to check it against, so the anchors ship as the questionnaire states them.
nguyen_2026_autonomy_student_voice	Item text is the English questionnaire the author deposited with the data; the survey of Hanoi lower-secondary students was very likely administered in Vietnamese, and that version is not in the deposit. Separately, the constructs in this deposit fall into two blocks that correlate negatively with each other (academic pressure, academic motivation and family factors against school autonomy support, voice, choice and ownership), and some item pairs correlate in the direction opposite to their wording. That pattern would fit one block having been stored reverse to the questionnaire’s 1 = Strongly disagree to 5 = Strongly agree, but there is no article to check it against, so the anchors ship as the questionnaire states them.
nguyen_2026_gad7	The item and option text are the English labels the authors deposited with the data. The survey of women at a Hanoi hospital was very likely administered in Vietnamese, and that version is not in the deposit, so the text shown is not confirmed to be what respondents read.
nguyen_2026_misfit_academic_performance	The deposit holds only English; respondents were undergraduates in Vietnam, so the survey was most likely administered in Vietnamese, which is not published, and the item text is the study’s own English. The questionnaire lists four academic-performance items but the data contain three (PE1-PE3). The item text follows the questionnaire’s codes, and the source does not say which item was dropped.
nguyen_2026_misfit_learning_misfit	The deposit holds only English; respondents were undergraduates in Vietnam, so the survey was most likely administered in Vietnamese, which is not published, and the item text is the study’s own English.
nguyen_2026_misfit_learning_motivation	The deposit holds only English; respondents were undergraduates in Vietnam, so the survey was most likely administered in Vietnamese, which is not published, and the item text is the study’s own English.
nguyen_2026_misfit_learning_satisfaction	The deposit holds only English; respondents were undergraduates in Vietnam, so the survey was most likely administered in Vietnamese, which is not published, and the item text is the study’s own English. The items are worded as dissatisfaction statements, so higher responses mean greater dissatisfaction with technology-enhanced learning, despite the table’s ‘learning satisfaction’ name.
nguyen_2026_misfit_technostress	The deposit holds only English; respondents were undergraduates in Vietnam, so the survey was most likely administered in Vietnamese, which is not published, and the item text is the study’s own English.
nguyen_2026_mspss	The item and option text are the study’s own English labels from its data file. The scale was administered in Vietnam and the Vietnamese wording respondents read was not deposited. The labels are not Zimet et al.’s (1988) standard English: item 5 reads ‘There is a special person who is…’, item 10 omits ‘in my life’, and the anchors read ‘Extremely’, ‘Strongly’ and ‘Moderately’ rather than ‘Very strongly’, ‘Strongly’ and ‘Mildly’.
nguyen_2026_sdt_academic_motivation	Item text is the English questionnaire the author deposited with the data; the Google Forms survey of Vietnamese university students was very likely administered in Vietnamese, and that version is not in the deposit.
nguyen_2026_sdt_perceived_performance	The deposit publishes only an English questionnaire; this survey of Vietnamese university students was probably administered in Vietnamese, and that wording is not available.
ni_2025_knowledge_transfer	The questionnaire was administered in Chinese to managers of Chinese SMEs, but neither the article nor its S1 Data publishes any Chinese wording, so the item text shown is the English the authors print in section 3.2; the assignment of items to KT-1..KT-4 follows the paper’s listing order, which the published factor loadings (0.919/0.925/0.951/0.931, reproduced from the live data as 0.9189/0.9247/0.9508/0.9310) pin to the column order but not to the wording order, and the paper publishes no labels for the five Likert points, so option_text is blank.
ni_2025_open_innovation	The item text shipped for this table is the English wording the source paper prints in its Methods; the questionnaire was administered to participants in Chinese, and no Chinese wording is published in the paper or its S1 Data. The order of items within each of the two blocks (OI1-1..OI1-4 and OI2-1..OI2-4) is inferred from the paper’s presentation order rather than from an explicit item-code label.
ni_2025_relationship_network	The questionnaire was administered in Chinese to managers of Chinese SMEs, but neither the article nor its S1 Data publishes any Chinese wording, so the item text shown is the English the authors print in section 3.2 of the paper; the assignment of wording to SN1-1..SN2-4 follows the paper’s listing order, which the published factor loadings (reproduced from the live data to within rounding on all eight items) pin to the column order but not to the wording order, and the paper publishes no labels for the five Likert points, so option_text is blank.
ni_2025_strategic_orientation	The questionnaire was administered in Chinese to managers of Chinese SMEs, but neither the article nor its S1 Data supplement publishes any Chinese wording, so the item text shown here is the English the authors print in section 3.2 of the paper; the assignment of items to the codes SO3-1..SO3-3 follows the paper’s own listing order, which the published factor loadings (0.914/0.920/0.919, reproduced from the live data as 0.9144/0.9199/0.9189) pin to the column order but not to the wording order, and the paper publishes no labels for the five Likert points, so option_text is blank
niazi_2020_mfq	The item text here is the canonical MFQ30 wording (moralfoundations.org, MFQ30.doc, July 2008), not the study’s own variable labels: those labels are an abridgement, dropping the stem “Whether or not” throughout Part 1 and carrying the depositor’s scoring notes on two items. The labels are what tie code to text and they reconcile 32/32 to the sentences shipped here. Note also that MATH and GOOD are the MFQ30’s two catch items and belong to no foundation subscale, so a five-foundation score computed over all 32 items in this table would be wrong.
niazi_2020_mfq_stereotype	The item wording here is the standard self-report MFQ30 form, but this table is the study’s stereotype condition: respondents rated each statement as they predicted a typical member of the opposite gender would answer it. The authors’ reworded stereotype instructions are not published in the paper or its supporting information, so the instruction and section-prompt fields are blank rather than carrying the self-referential canonical prompts.
nomt_hooper_2024_study1	Study 1 randomly assigned participants to an English- or Spanish-language version of the same task (63 English, 56 Spanish); the item text shipped here is the English wording, and the IRW response table does not record which language a given participant received.
nomt_hooper_2024_study2	A visual memory test using novel, unfamiliar object categories. Participants first studied six target objects, then on each trial chose which of three objects shown was one of the targets. The item is the trial number and resp is whether the choice was correct; chance is one in three. There is no question wording because the stimuli are pictures, which are in the original deposit but not reproduced here. The study compared groups given English and Spanish instructions, but the trials themselves are nonverbal and identical across groups.
number_pattern_game	Each item is one stimulus set, not one question: participants judged 30 different candidate numbers against each set, but the IRW table drops the source file’s target column, so which number a given 0/1 response referred to is not recoverable. item_text therefore ships the on- screen question template with [TARGET] in the slot where the number appeared.
odachi_2022_fear_covid19	The Japanese item text is the published Japanese FCV-19S (Wakashima et al. 2020), matched by item number, rather than wording supplied by this study. The English item and option wording in the _translated columns is the originators’ own (Ahorsu et al. 2020), but the English instruction line was written by IRW, because neither source publishes one.
offlinefriend_ofaq	Item text is transcribed verbatim from the questionnaire as administered (the study’s own OSF file), not from the companion scale document, because the administered form is what respondents read. The two disagree on three items, and the administered wording ships unchanged in each case including its typographical errors: item 18 reads “Arguments with arisen with others…” where the scale document and the paper print “have arisen”; item 33 lacks the scale document’s repeated “regularly”; item 35 reads “I had a strong craving” where the others print “I have”. Separately, the postprint’s Table 1 appears to misprint item 37’s mean as 2.82, a repeat of item 35’s printed value; the observed mean is 2.276, while its printed SD of 1.12 does reproduce (1.123).
ojelabi_2019_sf36	The item text is the canonical RAND 36-Item Health Survey 1.0 wording, transcribed from RAND’s own instrument page, not from the study: the article publishes no item wording and the PLOS S1 workbook’s headers are bare codes sf1..sf36 with no variable labels. Which canonical item each code carries is therefore an IRW reconstruction rather than anything the source states. It is verified two ways – per-item response ranges reproduce the SF-36’s block structure exactly (0 out-of-range responses, while all 35 cyclic shifts of the mapping produce 2-23 violations), and the workbook’s own SF-6D health-state code is an exact function of the SF-6D input items under this numbering – but a reader should know it was solved rather than read off a key. Two shipped option lists are one level wider than the sample used: sf21 never takes “Very severe” pain and sf2 never takes “Much worse”.
onah_2021_covid_knowledge	Respondents chose among lettered alternatives, which IRW has appended to each question in the item text, but the data record only whether the answer was scored correct (resp 1) or incorrect (0); the source does not publish which alternative was keyed correct. The instructions mention 4 marks per correct answer, but the table scores 1/0.
opladen2025_fkg	The English in this table’s item_text_translated column is the OSF deposit’s own short content gloss for each item (e.g. ‘joint pain as a sign of paralysis’), not a sentence-level translation; the German item_text is the wording respondents actually read. The deposit publishes only the abbreviation FKG, so the instrument name shipped here (Fragebogen zu Koerper und Gesundheit, Hiller et al. 1997) is an attribution from item content rather than a statement by the study, and the 20 items are a selection from that 53-item questionnaire
opladen2025_fks	The English in item_text_translated is the deposit’s own short content gloss per item (e.g. ‘body-related comparison’), not a sentence-level translation of the administered German, which is what item_text holds
opladen2025_wi	The English in item_text_translated is not a translation of the items: it is the deposit’s own short content gloss for each item (e.g. “worries about serious illness”). The German in item_text is the wording respondents read. The instrument is the 14-item Whiteley Index (Pilowsky, 1967) in German, not a body-image scale.
ordak_2026_vaccine_misreasoning	This is a content analysis, not a questionnaire: each id is a Polish-language anti-vaccination Facebook post and each item is one of ten statistical-misreasoning categories a coder applied to it, so item_text is the paper’s Table 1 definition of the category rather than wording anyone responded to. The response options were written by IRW (“Identified in the post” / “Not identified in the post”) because the paper publishes no option labels; resp 1 = identified, which reproduces the paper’s Table 2 counts for all ten categories.
orovou_2021_lec5	The study administered its own Greek translation of the LEC-5 and never published it; the English shipped is the wording in the study’s SPSS file, verbatim the National Center for PTSD’s LEC-5 Standard form. The response options are the study’s recoding, not the instrument’s: the LEC-5’s six exposure categories are collapsed to 0 = No and 1 = It happened to me/I was a witness, and no instructions are given because the published ones describe a format these respondents did not see.
orovou_2021_pcl5	The study administered a Greek translation of the PCL-5 but published no Greek wording, so the item text is the National Center for PTSD’s English original (verbatim against the standard form) rather than what respondents read. The paper’s Methods describes the scale as 1-5, while the deposit and the shipped anchors are the canonical 0-4 (0 = Not at all, 4 = Extremely). Instructions are blank because the reference window (past month vs past week) is not stated.
oxfordcovid_xue_2024_at	IRW missing the item: media_tv
ozkurt_2026_basic_psych_needs	The study administered a Turkish adaptation of the Basic Needs Satisfaction in Sport Scale and published no Turkish wording; the English here is the authors’ own rendering from their S6 File, which pairs one-for-one and in order with the 14 items of the original BNSSS (Ng, Lonsdale & Hodge, 2011) but diverges slightly from its wording (e.g. ‘I show interest in my teammates’ vs ‘I show concern for others in my sport’). Despite the dictionary description, this is the BNSSS, not Deci & Ryan’s Basic Psychological Need Satisfaction Scale.
ozkurt_2026_continuance_intention	The English is the study’s own rendering of a Turkish administration; the Turkish wording is not published. Items INT2 (‘I often think about quitting this sport.’) and INT3 (‘I intend to quit this sport.’) are stored reverse-scored, so for these two items resp 7 means strong disagreement with the printed wording. The paper’s Measures section cites a different, later five-point scale by the same author; the items here are the seven-point version listed in the study’s supplement.
ozkurt_2026_ego_orientation	Despite the table name, this is the complete 10-item goal orientation questionnaire: EGOQ1, 3, 4, 6 and 9 measure task orientation, not ego orientation. The study administered a Turkish adaptation (Ersoz et al., 2017) and published no Turkish wording; the English is the authors’ own rendering from their S6 File, which pairs in order with the original GOEM (Petherick & Markland, 2008) but recasts every item as a self-contained ‘I feel good when…’ sentence in place of the GOEM’s shared stem.
paampsmartsud_saba_2023_ders	The paper described only 36-item, but the IRW table contains 37
pakseresht_2021_gmfood_risk_ranking	The English is the authors’ own translation of a Swedish administration; the Swedish original is not in the article or any supplement. Respondents ranked the four risk dimensions after reading one of four randomly assigned GM-policy scenarios (Banned, R&D, Import, Full); that scenario text is not recorded in the response table, so the instructions carry a ‘<<[policy scenario]>>’ placeholder where each respondent saw a different description.
pang_2023_behavioral_intent	Item text is the study’s own English rendering; the questionnaire was administered in Chinese and no Chinese wording is published (the S2 File, despite its title, is English only). The paper’s Methods states 1 = strongly disagree to 5 = strongly agree, but the shipped anchors run the other way (1 = Definitely agree … 5 = Definitely disagree), following the questionnaire’s display order, which the deposit’s demographic codes corroborate – higher resp means MORE disagreement.
pang_2023_media_influence	Item text is the study’s own English rendering; the questionnaire was administered in Chinese and no Chinese wording is published (the S2 File, despite its title, is English only). The paper’s Methods states 1 = strongly disagree to 5 = strongly agree, but the shipped anchors run the other way (1 = Definitely agree … 5 = Definitely disagree), following the questionnaire’s display order, which the deposit’s demographic codes corroborate – higher resp means MORE disagreement.
pang_2023_perceived_cost	Item text is the study’s own English rendering; the questionnaire was administered in Chinese and no Chinese wording is published (the S2 File, despite its title, is English only). The paper’s Methods states 1 = strongly disagree to 5 = strongly agree, but the shipped anchors run the other way (1 = Definitely agree … 5 = Definitely disagree), following the questionnaire’s display order, which the deposit’s demographic codes corroborate – higher resp means MORE disagreement.
pang_2023_perceived_ease_use	Item text is the study’s own English rendering; the questionnaire was administered in Chinese and no Chinese wording is published (the S2 File, despite its title, is English only). The paper’s Methods states 1 = strongly disagree to 5 = strongly agree, but the shipped anchors run the other way (1 = Definitely agree … 5 = Definitely disagree), following the questionnaire’s display order, which the deposit’s demographic codes corroborate – higher resp means MORE disagreement.
pang_2023_perceived_enjoyment	Item text is the study’s own English rendering; the questionnaire was administered in Chinese and no Chinese wording is published (the S2 File, despite its title, is English only). The paper’s Methods states 1 = strongly disagree to 5 = strongly agree, but the shipped anchors run the other way (1 = Definitely agree … 5 = Definitely disagree), following the questionnaire’s display order, which the deposit’s demographic codes corroborate – higher resp means MORE disagreement.
pang_2023_perceived_risk	Item text is the study’s own English rendering; the questionnaire was administered in Chinese and no Chinese wording is published (the S2 File, despite its title, is English only). The paper’s Methods states 1 = strongly disagree to 5 = strongly agree, but the shipped anchors run the other way (1 = Definitely agree … 5 = Definitely disagree), following the questionnaire’s display order, which the deposit’s demographic codes corroborate – higher resp means MORE disagreement, i.e. less perceived risk.
pang_2023_perceived_usefulness	Item text is the study’s own English rendering; the questionnaire was administered in Chinese and no Chinese wording is published (the S2 File, despite its title, is English only). The paper’s Methods states 1 = strongly disagree to 5 = strongly agree, but the shipped anchors run the other way (1 = Definitely agree … 5 = Definitely disagree), following the questionnaire’s display order, which the deposit’s demographic codes corroborate – higher resp means MORE disagreement.
pang_2023_social_norms	Item text is the study’s own English rendering; the questionnaire was administered in Chinese and no Chinese wording is published (the S2 File, despite its title, is English only). The paper’s Methods states 1 = strongly disagree to 5 = strongly agree, but the shipped anchors run the other way (1 = Definitely agree … 5 = Definitely disagree), following the questionnaire’s display order, which the deposit’s demographic codes corroborate – higher resp means MORE disagreement.
papp_silva-martins2023	The item text here is the administered European Portuguese form. The English in item_text_translated is not a translation of it but O’Connor et al.’s (2014) original English PPAPP wording for the same numbered items, and three items diverge from the Portuguese: item 9 (Portuguese ‘futebol ou basquetebol’, English ‘soccer or baseball’), item 13 (the Portuguese names lessons, practices or games) and item 16 (the Portuguese switches to past tense, ‘how many times did you fail to enrol’). The English instructions and the five response-option labels were translated from the Portuguese by IRW; O’Connor’s own anchors differ (‘often’ where the Portuguese has ‘almost always’). Despite the table name, the instrument is the PPAPP, not the PAPP item bank.
paraguay_2021_mobility_mobility	The English wording for this table was produced by IRW, not by INE Paraguay, which publishes the AMA 2021 mobility survey in Spanish only; the No sabe (don’t know) option named in the instructions was offered to respondents but is recoded to missing in the response data.
paraguay_2021_women_childhood	The English wording for this table was produced by IRW, not by INE Paraguay, which publishes the ENSIMUP 2021 in Spanish only; the response codes were reordered in processing so that 1 = Nunca (never), 2 = De vez en cuando (now and then), 3 = Con mucha frecuencia (very often), unlike the questionnaire’s own 1 = De vez en cuando, 2 = Con mucha frecuencia, 3 = Nunca, and ‘don’t know/don’t remember’ answers are recoded to missing.
paraguay_2021_women_control	The English wording for this table was produced by IRW, not by INE Paraguay, which publishes the ENSIMUP 2021 in Spanish only; the item text is the current-partner question (1002, present tense), but the item codes also carry the answers of women without a current partner, who were asked the same six questions in the past tense about their ex-husband, ex-partner or last partner (question 1005, about a quarter of respondents); and the response codes were reversed in processing so that 1 = No and 2 = Si (yes), unlike the questionnaire’s 1 = Si, 6 = No.
paraguay_2021_women_family	The English wording for this table was produced by IRW, not by INE Paraguay, which publishes the ENSIMUP 2021 in Spanish only; the response codes were reversed in processing so that 1 = No and 2 = Si (yes), unlike the questionnaire’s 1 = Si, 6 = No.
paraguay_2021_women_justification	The English wording for this table was produced by IRW, not by INE Paraguay, which publishes the ENSIMUP 2021 in Spanish only; the response codes were reversed in processing so that 1 = No and 2 = Si (yes), unlike the questionnaire’s 1 = Si, 6 = No.
paraguay_2021_women_laws	The English wording for this table was produced by IRW, not by INE Paraguay, which publishes the ENSIMUP 2021 in Spanish only; these questions were asked only of women who first said they knew that laws protecting women’s rights exist (question 1204); and the response codes were reversed in processing so that 1 = No and 2 = Si (yes), unlike the questionnaire’s 1 = Si, 6 = No.
paraguay_2021_women_partner	The English wording for this table was produced by IRW, not by INE Paraguay, which publishes the ENSIMUP 2021 in Spanish only; women with no current partner (about a quarter of respondents) answered the past-tense version about their ex-husband, ex-partner or ex-boyfriend (question 1011), pooled under the same item codes; the item codes follow INE’s data dictionary, so from P1007_11 on they are one higher than the row numbers printed in the questionnaire (which omits a political-participation row, line 10, that is not in this table); and the response codes were reversed in processing so that 1 = No and 2 = Si (yes), unlike the questionnaire’s 1 = Si, 6 = No.
paraguay_2021_women_permission	The English wording for this table was produced by IRW, not by INE Paraguay, which publishes the ENSIMUP 2021 in Spanish only; item P1120F (decisions about her body and/or contraceptive methods) does not appear in the printed questionnaire and its wording comes from INE’s data dictionary only.
paraguay_2021_women_public	The English wording for this table was produced by IRW, not by INE Paraguay, which publishes the ENSIMUP 2021 in Spanish only; item codes follow INE’s data-dictionary line numbering (P601A_1, P601A_3 to P601A_22, no line 2), which runs one ahead of the printed questionnaire’s rows 2-21.
paraguay_2021_women_roles	The English wording for this table was produced by IRW, not by INE Paraguay, which publishes the ENSIMUP 2021 in Spanish only.
park_2024_ageism	Item text is the study’s own English questionnaire (S2 File); the scale was administered in Korean to Korean clinical nurses and no Korean wording is published. Responses are stored raw, so the four pro-elderly items (a4, a6, a11, a12) run opposite to the other fourteen; reverse them before scoring. The paper’s Methods describes the 29-item Fraboni Scale of Ageism, but the questionnaire and the deposit both hold 18 items, and the paper’s Table 2 statistics do not reproduce from the deposit.
pass20_klosowska_2025_dass_online	The 21-item Depression Anxiety Stress Scales, asking how much each statement applied to you over the past week, from 0 (did not apply to me at all) to 3 (applied to me very much or most of the time). The items are grouped into the scale’s three subscales of seven items each. The text is the English original – the study was run in Polish and that wording is not published in the deposit, so what is shown here is a rendering rather than what respondents read.
pauli_2021_coercion_attitudes	The study administered a German ad hoc translation of the Staff Attitude to Coercion Scale, which was never deposited, so the English wording from the study’s own codebook ships instead. The paper’s Methods describes a 1-6 scale running from agreement to disagreement, while the deposited responses are 0-5 with 5 = fully agree; the shipped anchors follow the deposit.
pavic_2022_healthcare_trust	The item text here is English, but the study was administered in Croatian to 577 college- and university-educated respondents in Croatia, and neither the paper nor its supplements publishes the Croatian wording, so these are not the sentences respondents read. The English is the study’s own, taken from the variable labels in its SPSS deposit, and matches 8 of the 9 items of Shea et al.’s (2008) revised Health Care System Distrust Scale word for word (Shea’s item 7, on equal treatment regardless of race or ethnicity, was dropped by the authors). Responses are stored raw: HCS_trust2, 4, 5, 7 and 8 are reverse-worded and must be reversed before a trust total is scored.
pavic_2022_natural_immunity	The item text here is English, but the study was administered in Croatian to 577 college- and university-educated respondents in Croatia (the instruments were translated and back-translated), and neither the paper nor its supplements publishes the Croatian wording, so these are not the sentences respondents read. The English is the study’s own, taken from the variable labels in its SPSS deposit, and matches the three natural-immunity items of Martin & Petrie’s (2017) published VAX scale verbatim.
pavic_2022_science_literacy	The item text here is English, but the study was administered in Croatian to 577 college- and university-educated respondents in Croatia, and neither the paper nor its supplements publishes the Croatian wording, so these are not the sentences respondents read. The English is the study’s own, taken from the variable labels in its SPSS deposit, and matches the 15 statements in the paper’s Table 4. The response levels are the study’s scoring, not the options respondents chose from: participants marked each statement true, untrue or don’t know, and the deposit keeps only the dichotomised result, so resp 0 pools incorrect and don’t-know answers. Table 4’s 72.77% correct for the genetically-modified-fruit item (Lit6) is not attainable over 577 respondents; the data give 72.27% (417 of 577).
pavic_2022_vaccine_conspiracy	The item text here is English, but the study was administered in Croatian to 577 college- and university-educated respondents in Croatia (the instruments were translated and back-translated), and neither the paper nor its supplements publishes the Croatian wording, so these are not the sentences respondents read. The English is the study’s own, taken from the variable labels in its SPSS deposit, and matches the seven stems of Shapiro et al.’s Vaccine Conspiracy Beliefs Scale as the paper prints them in Table 1.
pcl5_xiong_2025	The PCL-5 was administered in Chinese, but no Chinese wording is published; the item text is the English PCL-5 standard form from the US National Center for PTSD. Item codes pcl1-pcl20 are assumed to follow its numbering: the data confirm each item’s DSM-5 symptom cluster but not the order within a cluster.
pedroso_2021_ifsq_laissezfaire	Item text is the administered Brazilian-Portuguese IFSQ-Br wording, but the response anchors are English: the study publishes only the IFSQ’s English anchor sets, never the Portuguese respondents saw. Which of the two sets each item carried is inferred from item content – LF1-LF3 and LF6-LF9 are behaviour items (never … always), LF4, LF5, LF10 and LF11 belief items (disagree … agree). LF6recoded-LF9recoded are stored reverse-scored, so on those four resp 1 means “always” and 5 means “never”, the opposite of every other item. The English item wording is the study’s own back-translation (BT2).
pedroso_2021_ifsq_pressuring	The item text here is the administered Portuguese, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. Items embed the placeholder (nome da crianca), kept verbatim in both columns.
pedroso_2021_ifsq_responsiveness	Item text is the administered Brazilian-Portuguese IFSQ-Br wording, but the response anchors are English: the study publishes only the IFSQ’s English anchor sets, never the Portuguese respondents saw. Which of the two sets each item carried is inferred from item content – RP6, RP7 and RP12 are belief items (disagree … agree), the other nine behaviour items (never … always) – a split corroborated by those three being the only items with missing responses, since the study added a “don’t know” option to belief items only. The English item wording is the study’s own back-translation (BT2) of a pre-final draft and differs from the final Portuguese for RP1, RP2, RP8 and RP11.
pedroso_2021_ifsq_restriction	Item text is the administered Brazilian-Portuguese IFSQ-Br wording, but the response anchors are English: the study publishes only the IFSQ’s English anchor sets, never the Portuguese respondents saw. Which of the two sets each item carried is inferred from item content – RS1, RS2, RS5 and RS6 are behaviour items (never … always), RS3, RS4 and RS7-RS11 belief items (disagree … agree). RS5recoded and RS6recoded are stored reverse-scored, so on those two resp 1 means “always” and 5 means “never”, the opposite of every other item. The English item wording is the study’s own back-translation (BT2) of an earlier draft and lacks the parenthetical Brazilian food examples the final Portuguese adds to RS5-RS10.
pellerin2020_3dws	Item text for pellerin2020_3dws is the 3D-WS-12’s published English wording; the study administered the scale in French and no French version of the items appears in its OSF deposit or its article, so what is shown here is a parallel English version rather than the sentences respondents read. Within each facet the item-to-code order is inferred: the reflective block is verified against the instrument’s published loadings and freed residual, and the three positively-worded items land exactly where the authors’ own scoring key says they do, but Cog_1-Cog_4 could be any of the 24 orders of c9/c10/c11/c13 as far as the data can tell. Separately, the authors’ own wisdom score excludes Cog_1, whose responses correlate negatively with the reflective items once keyed in the same direction as its facet-mates (r = -0.21, -0.21, -0.21 against Refl_2/Refl_3/Refl_4 at wave 0, n=674, the wave the authors’ own score is computed on; -0.21 to -0.23 pooling all 1010 person-wave records); the IRW table keeps it.
pellerin2020_asti	Item text for pellerin2020_asti is the ASTI’s published English wording; the study administered the scale in French and no French version of the items or of the 1-7 anchors appears in its OSF deposit or its article, so what is shown here is a parallel English version rather than the sentences respondents actually read. The item-to-code mapping is a presentation-order inference that the response data confirms only in part: it ranks first of all 5040 possible assignments, and it fixes ST_4, ST_6 and ST_7, but it cannot distinguish ST_1 from ST_3 (‘I feel that my individual life is a part of a greater whole’ vs ‘I feel part of something greater than myself’, which correlate 0.76) or ST_2 from ST_5 (‘…belonging with both earlier and future generations’ vs ‘…oneness with nature’), so either of those two pairs could be swapped.
pellerin2020_cpc_hope	The item text here is the CPC-12’s published English (Lorenz et al., 2016, S1 Appendix), but the study was run in France and administered the scale in French; no French wording appears in the OSF deposit or the article, so the English stands in for what respondents read. Only the endpoint anchors (1 = strongly disagree, 7 = strongly agree) are published, as the article’s English gloss; points 2-6 are unlabelled. The deposit’s columns carry no labels, so Hope_1..Hope_3 are matched to CPC-12 hope items 1-3 by instrument order.
pellerin2020_cpc_optimism	The item text here is the CPC-12’s published English (Lorenz et al., 2016, S1 Appendix), but the study was run in France and administered the scale in French; no French wording appears in the OSF deposit or the article, so the English stands in for what respondents read, and the 1 = strongly disagree / 7 = strongly agree anchors are likewise the article’s English gloss. The mapping of Opt_1..Opt_3 onto CPC-12 items 4-6 is inferred from item order and only partly verified: Opt_1 is pinned by its age correlation (only item 4 mentions “the life ahead of me”), but Opt_2 and Opt_3 are near-synonymous and are separated only by an assumption about parent-scale cohesion.
pellerin2020_cpc_selfefficacy	The item text here is the CPC-12’s published English (Lorenz et al., 2016, S1 Appendix), but the study was run in France and administered the scale in French; no French wording appears in the OSF deposit or the article, so the English stands in for what respondents read, and the 1 = strongly disagree / 7 = strongly agree anchors are likewise the article’s English gloss. The mapping of SAE_1..SAE_3 onto CPC-12 items 10-12 is inferred from item order: age gradients, endorsement levels and well-being correlations all pick this assignment, but the items’ intercorrelation ordering does not replicate the CPC-12’s development samples, so it is not fully verified.
pellerin2020_gq6	The item text here is the GQ-6’s original published English (McCullough, Emmons & Tsang, 2002, Appendix A), but the study administered the French GQ-6 of Shankland & Martin-Krumm (2012) to a French sample; no French wording appears in the OSF deposit or the article, so the English stands in for what respondents read. The article prints only the endpoints (1 = strongly disagree, 7 = strongly agree); the five intermediate labels come from the GQ-6’s published form, not from the study. GQ_1..GQ_6 are mapped to GQ-6 items 1-6 by order, verified only at block level: GQ_3 and GQ_6 are confirmed as the reverse-keyed pair, but nothing distinguishes GQ_1 from GQ_2 or GQ_4 from GQ_5.
PEMAIW_Qiu_2020_PANAS	PANAS short form, twenty adjectives rated 1 = Very slightly or not at all to 5 = Extremely for how the respondent felt over the past week. Items 1, 3, 5, 9, 10, 12, 14, 16, 17 and 19 form the Positive Affect score and the other ten the Negative Affect score.
penningroth_2019_pm_concerns	The 15 items are content categories, not questions, and resp is a derived value rather than a response to the item shown: each participant listed up to five real-life prospective-memory tasks and marked, per task, which categories it related to as a concern, and resp 1 means at least one of the five tasks was concern-related in that category, 0 that none were. option_text describes that collapse rather than a printed option.
penningroth_2019_pm_goals	The 15 items are content categories, not questions, and resp is a derived value rather than a response to the item shown: each participant listed up to five real-life prospective-memory tasks and judged each against all 15 goal categories, and resp 1 means at least one of the five tasks was related to that goal category, 0 that none were.
perceived_injustice_mourin	Only 10 of this table’s 19 items (the retained final scale) carry wording, because the authors never published the other 9 draft items; the match of the 10 wordings to item codes assumes the final scale kept the draft’s item order, which the data confirm as a set but cannot check position by position.
perez_hernandez_2021_cesd7	As for perez_hernandez_2021_gad7: IRW’s English translation of the Spanish survey labels; the truncated cesd1 Spanish stem was completed by IRW with the CES-D standard wording.
perez_hernandez_2021_elcsa	As for perez_hernandez_2021_gad7: IRW’s English translation of the Spanish survey labels.
perez_hernandez_2021_gad7	The English item and option text (*_translated) was written by IRW from the Spanish survey labels. Two Spanish stems were truncated in the Stata labels (gad5 here, cesd1 in perez_hernandez_2021_cesd7) and were completed by IRW with the instrument’s standard wording.
personality_handwriting__financial_behavior	The 12 items are the Neuroticism block of this study’s questionnaire and mix financial-attitude with neuroticism personality statements, so the table name is narrower than its content; item_text is identical to item because the IRW item codes are the source spreadsheet’s own header sentences.
personalitychange_kramer_2025_rses	The study codebook labels response 5 ‘strongly disagree’, a duplicate of the response-1 label; IRW ships it as ‘strongly agree’, which the data confirm (the positively worded items correlate +0.59 to +0.62 with the same respondents’ life-satisfaction scores in all three studies). Responses 2 and 4 carry no label in the source, and the negatively worded items are stored as recorded, not reverse-scored.
personalitychange_kramer_2025_sccs	The study codebook labels response 5 ‘strongly disagree’, a duplicate of the response-1 label; IRW ships it as ‘strongly agree’, which the data confirm (the ten low-clarity items correlate -0.18 to -0.34 with the same respondents’ life-satisfaction scores, and sc01_11 ‘clear sense of who I am’ +0.32 to +0.38). Responses 2 and 4 carry no label in the source.
peters_2025_att_exp_eval	These are bipolar semantic-differential items: all 15 share the stem ‘I prefer feeling …’ and differ only in their anchor pair, so the item’s content sits in option_text at resp=1 and resp=max; intermediate points were not verbally labelled. The 14 seven-point items were displayed as -3..+3 (stored 1..7, so 4 is the midpoint) and ‘weird’ was a five-point 1..5 item. The source defines a 16th item (embarrassed) that has no responses in this table. Respondents answered in 21 languages (listed in the language column); only the English source wording is shipped, so most non-English respondents read a translation of it, and each respondent’s language is in cov_language.
peters_2025_att_exp_strength	This table pools the Your COVID-19 Risk tool’s deployments in about 20 languages across 21 countries, but only the English master wording from the study’s question-definition sheet ships, so respondents outside the English deployments read a translation of it; each respondent’s language is in cov_language. All 16 items share one identical stem and differ only in their anchor pair, so each item’s content is in option_text at resp 1 and resp max, not in item_text. The 7-point items were displayed as -3..+3 and are stored 1..7 (stored 4 is the midpoint); “weird” is a 1..5 item.
peters_2025_att_instr_eval	This table pools the Your COVID-19 Risk tool’s deployments in about 22 languages across 21 countries, but only the English master wording from the study’s question-definition sheet ships, so respondents outside the English deployments read a translation of it; each respondent’s language is in cov_language. The bipolar items were displayed as -3..+3 and are stored 1..7, with wording only at the two poles.
peters_2025_att_instr_strength	The Your COVID-19 Risk tool was fielded in 22 languages and this table pools all of them, but IRW ships only the English master wording from the study’s question-definition spreadsheet, so respondents outside the English deployments read a translation of it (per-respondent language is in cov_language). In every item {{distRec}} stands for the respondent’s country’s recommended distance (1 metre to 2.5 metres, or 6 feet), and the bipolar items were displayed as -3 to +3 but are stored as 1 to 7, with wording only at the two poles.
peters_2025_ciberlite	The English master wording is shipped for a survey administered in 21 languages (about 55% of responses were to a non-English version); ‘{{distRec}}’ stands for each country’s recommended physical distance, and items pbccap1 and pbccap2 share an identical English stem, differing only in their response anchors.
peters_2025_gen_risk_eval	Both items share the evaluation stem ‘I prefer to (get) …’ and are completed by their endpoint anchors (resp=1 ‘certainly not end up in a hospital’ / ‘no symptoms’, resp=5 ‘certainly end up in a hospital’ / ‘severe symptoms’); intermediate points were not verbally labelled. The source defines a third item (dying) that has no responses in this table, and countries that received the amended wording (‘For me, avoiding … is …’) are not included. Respondents answered in 15 languages (listed in the language column); only the English source wording is shipped, so most non-English respondents read a translation of it, and each respondent’s language is in cov_language.
peters_2025_gen_risk_strength	This table pools the Your COVID-19 Risk tool’s deployments in many languages across 21 countries, but only the English master wording from the study’s question-definition sheet ships, so respondents outside the English deployments read a translation of it; each respondent’s language is in cov_language. Only the endpoints of the 1-5 scale are labelled.
peters_2025_gnrc_beliefs	This table pools the Your COVID-19 Risk tool’s deployments in about 20 languages across 21 countries, but only the English master wording from the study’s question-definition sheet ships, so respondents outside the English deployments read a translation of it; each respondent’s language is in cov_language. Only the endpoints of the 1-5 scale are labelled.
peters_2025_hw_frequency	These seven items are the tick-boxes of one select-all-that-apply question (‘Do you regularly wash or clean your hands in the following situations?’): resp 1 means the box was ticked and resp 0 that it was not, which also covers respondents who skipped the question, and the boxes carried no response labels, so option_text is blank. The tool was fielded in 22 languages and only the English master wording is shipped (per-respondent language is in cov_language); the English original misspells ‘someone’ as ‘somone’ in item cntct, kept as printed.
peters_2025_hw_intensity	This is a single check-all-that-apply question (‘If you wash/clean your hands, how do you do this?’): each item is one checkbox option, resp=1 means the option was checked and resp=0 that it was not, so a respondent who skipped the question is coded 0 on all four (540 of 70,066 respondents) and ‘I don’t do any of the above’ was not enforced as exclusive (48 respondents checked it alongside another option). Respondents answered in 22 languages (listed in the language column); only the English master wording is shipped, and each respondent’s language is in cov_language.
peters_2025_nrm_desc_behavior	These items were answered on a five-point scale labelled only at its ends (‘very improbable’ = 1, ‘very probable’ = 5); ‘{{distRec}}’ in each stem is the study’s own placeholder for the country’s recommended physical distance. Respondents answered in 20 languages (listed in the language column); only the English source wording is shipped, so most non-English respondents read a translation of it, and each respondent’s language is in cov_language.
peters_2025_nrm_desc_ident	The Your COVID-19 Risk tool was fielded in 20 languages (listed in the language column) and this table pools all of them, but IRW ships only the English master wording from the study’s own question-definition sheet, so non-English respondents read a translation of it; each respondent’s language is in cov_language. The bipolar items were displayed as -3 to +3 and are stored as 1 to 7 (stored 4 is the midpoint), with wording only at the two poles, and ‘{{distRec}}’ in the stem is the study’s own placeholder for the country’s recommended physical distance.
peters_2025_nrm_inj_approval	This table pools the Your COVID-19 Risk tool’s deployments in about 20 languages across 21 countries, but only the English master wording from the study’s question-definition sheet ships, so respondents outside the English deployments read a translation of it; each respondent’s language is in cov_language. The bipolar items were displayed as -3..+3 and are stored 1..7 (stored 4 is the midpoint), with wording only at the poles, and “{{distRec}}” is the study’s placeholder for the country’s recommended physical distance.
peters_2025_nrm_inj_motivation	The Your COVID-19 Risk tool was fielded in 22 languages (listed in the language column, most-answered first) and IRW ships only the English master wording from the study’s own question-definition spreadsheet, so most non-English respondents read a translation of it; each respondent’s language is in cov_language. Only the two endpoints of the 1-5 scale were labelled (‘not at all’, ‘very much’), and ‘{{distRec}}’ in the stem is the study’s own placeholder for the country’s recommended physical distance.
peters_2025_pbc_cond_power	This table pools the Your COVID-19 Risk tool’s deployments in about 20 languages across 21 countries, but only the English master wording from the study’s question-definition sheet ships, so respondents outside the English deployments read a translation of it; each respondent’s language is in cov_language. The bipolar items were displayed as -3..+3 and are stored 1..7, with wording only at the poles (“much harder” at 1, “much easier” at 7), and “{{distRec}}” is the study’s placeholder for the country’s recommended physical distance.
peters_2025_pbc_cond_presence	This table pools the Your COVID-19 Risk tool’s deployments in about 22 languages across 21 countries, but only the English master wording from the study’s question-definition sheet ships, so respondents outside the English deployments read a translation of it; each respondent’s language is in cov_language. “{{distRec}}” is the study’s placeholder for the country’s recommended physical distance, and only the endpoints of the 1-5 scale (“very unlikely” / “very likely”) are labelled.
peters_2025_pbc_skill_import	The Your COVID-19 Risk tool was fielded in 20 languages (listed in the language column, most-answered first) and IRW ships only the English master wording from the study’s own question-definition spreadsheet, so most non-English respondents read a translation of it; each respondent’s language is in cov_language. Only the two endpoints of the 1-5 scale were labelled (‘not at all important’, ‘extremely important’), ‘{{distRec}}’ in each stem is the study’s own placeholder for the country’s recommended physical distance, and item remem is shipped with the source’s missing word (‘remembering to do is …’, meaning ‘remembering to do so’).
peters_2025_pbc_skill_prob	The Your COVID-19 Risk tool was fielded in 20 languages (listed in the language column) and this table pools all of them, but IRW ships only the English master wording from the study’s own question-definition sheet, so non-English respondents read a translation of it; each respondent’s language is in cov_language. Only the two endpoints of the 1-5 scale were labelled (‘not able at all’, ‘absolutely able’); ‘{{distRec}}’ in each stem is the study’s own placeholder for the country’s recommended physical distance; and the deccont stem reads ‘decline physical content’ exactly as the source printed it, where ‘contact’ is evidently meant.
peters_2025_si_current	These six items are the tick-boxes of one select-all-that-apply question (‘Are you doing any of the following?’): resp 1 means the box was ticked and resp 0 that it was not, which also covers respondents who skipped the question, and the boxes carried no response labels, so option_text is blank; ‘Mostly staying at home’ was not exclusive, and most who ticked it also ticked a reason for leaving the house. The tool was fielded in 22 languages and only the English master wording is shipped (per-respondent language is in cov_language).
peters_2025_si_intention	These five items are the tick-boxes of one select-all-that-apply question (‘What would change?’, asked after ‘I would leave my house less (or not at all) if:’): resp 1 means ticked and resp 0 not ticked, and the boxes carried no response labels, so option_text is blank; each option was shown only to respondents who had ticked the matching behaviour in peters_2025_si_current, but everyone has a row for all five, so resp 0 also covers respondents who never saw the option (for example 47414 of 73949 for ‘I would stop visiting family’) and should be read conditional on peters_2025_si_current. The tool was fielded in 22 languages and only the English master wording is shipped (per-respondent language is in cov_language).
peters_2025_si_trigger	These four items are the tick-boxes of one select-all-that-apply question (‘I would leave my house less (or not at all) if:’): resp 1 means the box was ticked and resp 0 that it was not, which also covers respondents who skipped the question, and the boxes carried no response labels, so option_text is blank; ‘I would not consider this at all’ was not exclusive, and 645 of the 3385 who ticked it also ticked a trigger, while the question’s free-text ‘Other’ box is not included. The tool was fielded in 22 languages and only the English master wording is shipped (per-respondent language is in cov_language).
peters_2025_work_precautions	These four items are the tick-boxes of one select-all-that-apply question (‘Currently, do you work within a short distance of others?’): resp 1 means the box was ticked and resp 0 that it was not, which also covers respondents who skipped the question, and the boxes carried no response labels, so option_text is blank; ‘I am working from home or not working currently’ was not exclusive, and 1430 respondents ticked it together with a ‘Yes, with …’ option. The tool was fielded in 22 languages and only the English master wording is shipped (per-respondent language is in cov_language).
phq_BrummerHoffman_2021	This table pools two administration arms whose wording differed by design: the text shown is the Study 1 grid form (the shared frame in section_prompt, the symptom in item_text), while Study 2 respondents saw each item as one complete Portuguese sentence, e.g. ‘Nas ultimas duas semanas, em quantos dias voce teve pouco interesse ou prazer em fazer as coisas?’. The English translated columns are mixed provenance – seven symptom phrases come from the deposit’s own English questionnaire, but the frame stems, all anchors and the insomnia and suicide rows were translated for IRW, because the deposit’s English form prints canonical PHQ-9 anchors and an oversleeping clause that were not administered.
phq9_xiong_2025	The PHQ-9 was administered in Chinese, but no Chinese wording is published in this study’s data deposit, so the item and option text is the publisher’s official English (phqscreeners.com) rather than the sentences respondents read. Item order is assumed to follow the standard PHQ-9 numbering. The table pools adolescents, college students, community adults and male prisoners with no covariate identifying the sample.
pickova2025_moral_justifications	The instructions stem is shipped exactly as the study’s Google Forms export records it: “Please indicate how much you 6 with the following statements…”. The “6” is an artifact in the source file where “agree” belongs; respondents almost certainly read “agree”, and IRW has not repaired it. The 1-7 scale has no published anchor labels, so option_text is blank and the direction of the scale is not established.
pierro_2018_assessment_s1	The item text here is the original English Regulatory Mode Questionnaire (Kruglanski et al., 2000), but Study 1 administered an Italian version to 323 Italian adults, and neither the PLOS ONE article nor its SPSS deposit publishes the Italian wording, so these are not the sentences respondents read. The three reverse-worded items (assR1, assR5, assR11) are stored already reverse-recoded, so their anchors are flipped: for these, resp 1 = Strongly Agree and resp 6 = Strongly Disagree with the printed statement. Only the endpoints of the 6-point scale are labelled, following the paper’s description of the administration.
pierro_2018_assessment_s3	The item text here is the original English Regulatory Mode Questionnaire (Kruglanski et al., 2000), but Study 3 administered the same Italian version used in Study 1, and neither the PLOS ONE article nor its SPSS deposit publishes the Italian wording, so these are not the sentences respondents read. The deposit’s columns carry no labels, so the wording is matched to the codes by the instrument’s subscale order and the positions of the reverse-worded items (1, 5 and 11). Unlike Study 1, those three items (assr1, assr5, assr11) are stored raw here, not reverse-recoded.
pierro_2018_assessment_s4	The three reverse-worded items (assR1, assR5, assR11) are stored already reverse-recoded, so their shipped anchors are flipped: for these items resp=1 is ‘Strongly Agree’ and resp=6 is ‘Strongly Disagree’ with the printed statement. Item wording is the original English Regulatory Mode Questionnaire; the study administered an Italian translation whose wording is not published in the deposit.
pierro_2018_locomotion_s1	The item text here is the original English Regulatory Mode Questionnaire (Kruglanski et al., 2000), but Study 1 administered an Italian version to 323 Italian adults, and neither the PLOS ONE article nor its SPSS deposit publishes the Italian wording, so these are not the sentences respondents read. The two reverse-worded items (locaR6, locaR9) are stored already reverse-recoded, so their anchors are flipped: for these, resp 1 = Strongly Agree and resp 6 = Strongly Disagree with the printed statement. Only the endpoints of the 6-point scale are labelled, following the paper’s description of the administration.
pierro_2018_locomotion_s3	The item text here is the original English Regulatory Mode Questionnaire (Kruglanski et al., 2000), but Study 3 administered the same Italian version used in Study 1 to 85 Italian participants, and neither the PLOS ONE article nor its SPSS deposit publishes the Italian wording, so these are not the sentences respondents read. Unlike the Study 1 table, the two reverse-worded items (locr6, locr9) are stored raw here, so a high response on them means the reverse-worded statement is true of the respondent and must be reversed before scoring. Only the endpoints of the 6-point scale are labelled, following the paper’s description.
pierro_2018_locomotion_s4	The two reverse-worded items (locR6, locR9) are stored already reverse-recoded, so their shipped anchors are flipped: for these items resp=1 is ‘Strongly Agree’ and resp=6 is ‘Strongly Disagree’ with the printed statement. Note this differs from the sibling pierro_2018_locomotion_s3, where the same two items are stored raw. Item wording is the original English Regulatory Mode Questionnaire; the study administered an Italian translation whose wording is not published in the deposit. Only the two endpoints of the 6-point scale are labelled, following the paper’s own description; resp 2-5 carry no option text.
pierro_2018_selfforgive_s1	The item text here is the English wording printed in the study’s own PLOS ONE article (its rendering of a 4-item adaptation of Wohl et al.’s State Self-Forgiveness Scale), but the scale was administered in Italian to 323 Italian adults and neither the article nor its SPSS deposit publishes the Italian form, so these are not the sentences respondents read. The two reverse-worded items (sforgiveR1 ‘rejecting of myself’, sforgiveR3 ‘dislike toward myself’) are stored already reverse-recoded, so their anchors are flipped: for these, resp 1 = Completely and resp 4 = Not at all. Only the endpoints of the 4-point scale are labelled.
pierro_2018_selfforgive_s2	The item text here is the English wording printed in the study’s own PLOS ONE article (its rendering of a 4-item adaptation of Wohl et al.’s State Self-Forgiveness Scale), but the scale was administered in Italian to 97 Italian university students and neither the article nor its SPSS deposit publishes the Italian form, so these are not the sentences respondents read. The two reverse-worded items (SforgiveR1 ‘rejecting of myself’, SforgiveR3 ‘dislike toward myself’) are stored already reverse-recoded, so their anchors are flipped: for these, resp 1 = Completely and resp 4 = Not at all. Only the endpoints of the 4-point scale are labelled.
pierro_2018_selfforgive_s3	The item text here is the English wording printed in the study’s own PLOS ONE article (its rendering of a 4-item adaptation of Wohl et al.’s State Self-Forgiveness Scale), but the scale was administered in Italian to 85 Italian respondents and neither the article nor its SPSS deposit publishes the Italian form, so these are not the sentences respondents read. Unlike the sibling tables pierro_2018_selfforgive_s1 and _s2, the two reverse-worded items (sforgiver1 “rejecting of myself”, sforgiver3 “dislike toward myself”) are stored RAW here, not reverse-recoded, so their anchors read as printed; this was confirmed by reproducing the paper’s published composite. Only the endpoints of the 4-point scale are labelled, so option_text is blank for resp 2 and 3. The stem “I feel…” is printed once in the paper and is repeated on each item row here so every item carries its own referent.
pierro_2018_tf_future_s4	The item text here is the Temporal Focus Scale’s original English (Shipp, Edwards & Lambert, 2009), but Study 4 was run in Italian and neither the PLOS ONE article nor its SPSS deposit publishes the Italian form, so these are not the sentences respondents read. The deposit’s columns carry no labels, and TFfuture1-4 are assigned to the four future items in the instrument’s published order; the data cannot confirm this, since the composite reproduces the paper’s statistics under any reordering. Only resp 7 (Constantly) is labelled; the paper’s other endpoint, 1 = Never, is absent because no respondent used it.
pierro_2018_tf_past_s4	The item text here is the Temporal Focus Scale’s original English (Shipp, Edwards & Lambert, 2009), but Study 4 was run with Italian students and neither the PLOS ONE article nor its SPSS deposit publishes the Italian form, so these are not the sentences respondents read. The deposit’s columns carry no labels, and TFpast1-4 are assigned to the four past items in the instrument’s own order; the items are same-polarity near-synonyms, so a permutation would be undetectable. Treat the subscale as reliable and each sentence’s attachment to a code as provisional.
pierro_2018_tf_present_s4	Item text here is the Temporal Focus Scale’s original ENGLISH wording (Shipp, Edwards & Lambert 2009); the study was run on Italian students and neither the PLOS ONE article nor its SPSS deposit publishes the Italian form, so this is a same-instrument substitute rather than what respondents read. Separately, the assignment of the four sentences to the codes TFpresent1-TFpresent4 is inferred from the instrument’s own item order and cannot be checked against the data: the subscale identity is solid (the composite reproduces the paper’s published Present statistics, M 5.115/SD 1.001/alpha .854 against 5.12/1.00/.85, and its entire Table 4 correlation row to within .004), but the four items are same-range, same-polarity near- synonyms, so a permutation among them would be undetectable. Treat the subscale as reliable and an individual sentence’s attachment to an individual code as provisional.
pietraszkiewicz_2017_leader_eval	The item wording is the English printed in the paper. The survey ran at a Catalan university and the paper does not state its language, so respondents may have read a Catalan or Spanish version. The paper quotes the aspect wording only from the leaders’ self-evaluation form and says team members rated ‘the same four aspects’, so the wording here is taken from the self-evaluation prompt; ‘[name of leader]’ stands in for the leader’s name.
pilch_2021_coping_behavior	The item text here is the authors’ own English for their coping-behaviour scale (from the paper’s S1 Table), but the survey was administered in Polish to Polish adults and no Polish wording appears in the deposit or supplements, so these are not the sentences respondents read. The deposit’s variable labels give each code’s subscale but no wording, so within each of the three subscales the statements are matched to codes by S1 Table’s listing order; the data confirm subscale membership only, not the order inside a subscale.
pilch_2021_fcv19s_validation	The item text is the Polish Fear of COVID-19 Scale as printed in the study’s own deposited questionnaire (the Materials sheet of the PeerJ supplement). Ahorsu et al.’s original English wording is in the _translated columns. For items 5-7 the deposited Polish differs from the translation printed in the paper’s Table 1: item 5 omits “w mediach społecznościowych” (on social media), and items 6 and 7 are phrased differently. The deposited version is shipped as the one respondents saw.
pilch_2021_fear_covid19	The item text here is the administered Polish Fear of COVID-19 Scale, taken from the Polish validation paper the study cites (Pilch, Kurasz & Turska-Kawa, 2021, Table 1). The English in the _translated columns is Ahorsu et al.’s original FCV-19S wording as that paper prints it, a parallel original rather than a translation of the Polish beside it. The Polish response anchors are not published, so option_text is blank and only the English endpoint labels (1 = strongly disagree, 5 = strongly agree) are carried in the _translated column.
pilch_2021_personality_ipip20	The item text here is the published Polish IPIP-BFM-20 form (Topolewska et al., 2014), not wording printed by this study, whose data file labels its columns only IPIP1-IPIP20. The English item statements are IPIP’s own wording for the same numbered items, a parallel original rather than a translation of the Polish, while the English of the instructions and the five response anchors was written by IRW. The instructions are the paper-and-pencil form’s (“mark an X in the box”), although this study ran online. Codes are matched to wording by the form’s item numbering; the data confirm each item’s trait and keying, but a swap between the two same-trait, same-direction items in a pair (e.g. IPIP1/IPIP11) would be invisible.
pilch_2021_protection_motivation	The item text here is the authors’ own English for their protection-motivation questionnaire (from the paper’s Table 1), but the survey was administered in Polish and no Polish wording appears in the deposit or supplements, so these are not the sentences respondents read. The deposit’s variable labels give each code’s subscale but no wording; within each two-item subscale, which statement is item 1 and which is item 2 follows Table 1’s listing order and is not independently verified.
pinheiro_2023_srq	The item text is the World Health Organization’s English SRQ-20 (WHO/MNH/PSF/94.8). Respondents answered the Brazilian Portuguese adaptation, which was not published with the study and whose wording can differ; for example, the paper renders item 9 as ‘Have you been feeling sad lately?’ where WHO’s English reads ‘Do you feel unhappy?’. Item order is assumed to follow the standard SRQ-20 numbering.
pks_probability	The item text is the English wording printed in the pks R package manual by the researchers who collected the data; the questionnaire was presumably administered in German (a University of Tuebingen sample using a 32-card deck and euro coins), and the German version is not published. Problem b110 was worded differently for the two groups of respondents: the 26 lab respondents (ids 1-26) were told the probability of drawing a red marble is 0.30 (correct answer 0.28), the 478 online respondents 0.70 (correct answer 0.12). The item text carries the manual’s combined sentence, and the table has no covariate for mode, so lab respondents can be identified only by id.
PMT_Trzcinska_2023_PMT	The Polish item text ships both gender-matched forms of every item in one field, marked [girls’ form] and [boys’ form], because the item code does not record which form a child saw. The 32 TWO* item codes are the same 32 questions re-administered two weeks later to a 27-child retest subsample, not new items. The English in the _translated columns was assembled by IRW from the paper’s own English wording of each picture and question; the study never prints them as sentences.
polca_carcinoma	Seven pathologists each rated the same 118 slides for whether carcinoma was present in the uterine cervix. The items here are the pathologists, labelled A to G, and the persons being measured are the slides – the reverse of a questionnaire. There is no question wording: each entry is one rater’s verdict on one slide. Responses are 0 for no and 1 for yes; the original source coded these 1 and 2, which is kept alongside.
political_psychology	Item 27 (2016 vote report) stores a merged category 3 = ‘did not vote’ or ‘planned to vote but forgot’ (voters for other candidates are missing), and item 28 is not a survey question but the respondent’s 2016 vote recorded by Prolific.
portella_2022_racial_attitudes	Item q9 is stored in the opposite direction to the other three (for q9, 1 = Strongly agree and 4 = Strongly disagree; for q12, q57 and q60, 1 = Strongly disagree and 4 = Strongly agree). Item wording is the study’s English rendering of a Portuguese questionnaire whose Portuguese statements were not published; instructions and anchors ship in the administered Portuguese.
powell_2018_qcae	The item wording is not from Powell’s own materials: the deposit states only that each column is a QCAE item number, so the text is the canonical QCAE keyed by that number, taken from an independent QCAE deposit (Gomez et al., 2022) for 27 items and from the published instrument for the four reverse-worded items 1, 2, 17 and 29. The two middle anchors are the canonical QCAE labels; the study publishes only the endpoints. The data confirm subscale membership exactly, but order within a subscale rests on the item numbering. Responses are stored raw, so items 1, 2, 17 and 29 must be reversed before scoring.
poza2026_hlseu	This table was administered in two languages – 93 of the 101 respondents answered the Arabic form and 8 the French form – so item_text carries both administered wordings in one field, marked [Arabic version] and [French version]. The English in the _translated columns was written by IRW, not by the study. The response labels shipped for 0 and 1 are the data file’s own Spanish labels for the study’s difficult/easy dichotomisation (‘DIFICIL/MUY DIFICIL’, ‘MUY FACIL/FACIL’), because the administered Arabic and French anchors are published nowhere.
preschool_sel_emt	For the twelve Part 1 items and emta3_1s_t1 the item text is the study’s SPSS variable label, which SPSS truncated at 255 characters, so those prompts end mid-sentence; the child- facing question in Part 1 is ‘Show me which one of these feels the same as this one’. Item text is the interviewer’s administration script, including instructions to point at the stimulus, rather than a cleaned item stem.
preussmattsson_2022_ownership	The 24 item codes are not 24 statements: they are the six body-ownership statements S1-S6 crossed with the four visuo-vestibular x visuo-tactile synchrony conditions, and the code’s suffix (AVVAVT, AVVSVT, SVVAVT, SVVSVT) names the condition. Each statement’s wording therefore appears on four codes, and responses to the same statement under different conditions are separate items. The statements were read from the paper’s Table 2 (an image). resp runs -3 to +3 and only the endpoints (“fully disagree”, “fully agree”) are published, so option_text is blank for the five middle points.
previc_bohn2023	This table holds the PREVIC’s initial 379-word calibration pool, not the 89-word published PREVIC; each item code is the German word on the card, and the per-word English glosses are the study authors’ own. The English instructions and the YES/NO option labels were translated from the German by IRW – the on-screen controls were an unlabelled check mark and cross. The German instructions come from the public 2025 demo of the task app, so minor wording drift from the 2021 administration cannot be excluded.
prihandoko_2025_ews	The English translations (item_text_translated) were written by IRW from the Indonesian statements respondents read; they are not the study’s own.
promis1wave1_cesd	The 20-item CES-D. resp runs 0-3 from ‘Rarely or none of the time (Less than 1 day)’ to ‘Most or all of the time (5-7 days)’, one shared response set across every item. Items 4, 8, 12 and 16 are positively worded and so score in reverse of the other sixteen.
promis1wave1_haq	The HAQ disability items. 20 items ask how much difficulty an activity caused over the past week, resp 0-3 from ‘Without any difficulty’ to ‘Unable to do’; the other 4 (HAQ9-HAQ12) ask whether an aid or another person is usually needed and are yes/no, resp 0-1. The numeric prefix marks the instrument’s activity category, so HAQ1a and HAQ1b are two items about the same category.
protestant_workethic	Only items 1-19 of this table are Protestant Work Ethic Scale items. Items 20-29 are the Ten Item Personality Inventory (1-7 scale) and items 30-45 are a 16-word vocabulary check-list (0 = unchecked, 1 = checked, including the non-words ‘cuivocal’, ‘florted’ and ‘verdid’ as a validity check), both administered after the PWE in the same session.
psychoneurotic_inventory	The instructions shipped with this table are the study site’s own wording and state that the inventory has 112 yes/no questions; the deposit and the response data both contain 116 items (1-116).
ptacek2023_dass21	The item text here is the Vilimovsky & Kucera Czech DASS-21 that the study names as its administered version, taken from the instrument’s distribution site because neither the OSF deposit nor the paper reproduces it. The English in the _translated columns is the publisher’s parallel English DASS-21, not a translation of the Czech, so some items do not correspond word for word (Czech item 8 reads “I felt that I was tense and nervous” where the English reads “I felt that I was using a lot of nervous energy”). The data confirm the depression/anxiety/stress partition, but order within a subscale rests on the code numbering.
purnama_2023_attitude	The item text is the English wording in the deposit’s codebook. The survey was given to residents of Denpasar, Bali; its language is not stated and no Indonesian wording is published.
purnama_2023_constraints	As for purnama_2023_attitude: the deposit’s English wording; the administered wording is not published.
purnama_2023_digital_need	As for purnama_2023_attitude: the deposit’s English wording; the administered wording is not published.
purnama_2023_environment	As for purnama_2023_attitude: the deposit’s English wording; the administered wording is not published.
purnama_2023_program_benefit	As for purnama_2023_attitude: the deposit’s English wording; the administered wording is not published.
purnama_2023_threat	As for purnama_2023_attitude: the deposit’s English wording; the administered wording is not published.
QADT_Kleka_2023	Twelve forced-choice items about attainment of developmental tasks in school-aged children, administered in Polish with the authors’ own English alongside. resp is 1 when the task is attained and 0 when it is not; six items were reverse-worded and the source has already recoded them, so the A/B statement attached to resp 1 differs by item. The items have no stems – the two statements are the whole item – and share one lead-in, ‘Zycie szkolne jest:’ (‘School life is:’).
QCDQES_Oliveira_2022	The Portuguese item, option and instruction wording here is the study’s own, transcribed from its SPSS value labels. The English in the _translated columns was produced by IRW – it is not the study’s wording and has not been reviewed or back-translated by a Portuguese speaker. Treat it as a reading aid rather than a validated English instrument.
qi_2024_mspss	As for qi_2024_phq9: the item text and options are the deposit’s own “English-language codebook”, not the Chinese wording respondents read. The 1-7 scale is labelled in the source at its endpoints only (“strongly disagree” / “strongly agree”), so codes 2-6 carry no option text rather than invented labels.
qi_2024_phq9	The item text and response options here come from the deposit’s own “English-language codebook”, which is the authors’ English rendering of a questionnaire that was administered to Chinese nurses in Chinese. No Chinese wording is published in the deposit or the article, so this is not the text respondents read. The mapping from item code to text is the codebook’s own and is exact; only the language is a substitute.
qi_2024_somatic	Two things. First, the source instrument has TWENTY items and this IRW table carries FIFTEEN: five columns in the deposit’s SSS block are mislabelled with other scales’ names (PHQ3.1, PHQ4.1, PHQ1.1, GAD1, PHQ7.1) and were excluded when the response table was built. They are the missing SSS items – the deposit’s own SSS Total score equals the sum of the fifteen shipped items plus those five for every one of the 293 respondents, and equals the fifteen-item sum for none – but which of items 16-20 each one is has not been established, so none of them appears here. Item text is shipped for the fifteen that are in the table. Second, as for qi_2024_phq9, that text is the deposit’s own “English-language codebook” rather than the Chinese wording respondents read.
qi_2025_dark_triad	Respondents completed the Chinese Short Dark Triad (SD3-C; Zhang, Ziegler & Paulhus, 2020), but its wording appears in neither the OSF deposit nor the supplement, so the item text is the SD3’s English original as distributed by its author (SD3.1.1) – a parallel version of the same numbered items, not what respondents read. The 1-5 anchors are the English instrument’s; the paper labels only the endpoints. Codes follow the SD3’s item order, which the published subscale alphas confirm at block level but not within a block. Responses are stored raw: DT_11, DT_15, DT_17, DT_20 and DT_25 are reverse-keyed and not yet reversed.
qi_2025_panas	The instrument is the Qiu, Zheng & Wang (2008) Chinese revision of the PANAS, not Watson et al.’s 20-item original: it has 18 items and six of them (grateful, joyful, elated and jittery among them) are not Watson adjectives. Neither the study’s OSF deposit nor its supplement contains any questionnaire wording, so the Chinese item, anchor and instruction text was taken from the Institute of Psychology (CAS) Cognitive Experiment Standard Toolbox’s own administration of that instrument and corroborated item-for-item against a second independent reproduction. The English in the _translated columns was produced by IRW, not by the study.
qi_2025_self_concept_clarity	The Self-Concept Clarity Scale was administered in Chinese, but its wording appears in neither the OSF deposit nor the supplement, so the item text is Campbell et al.’s (1996) English original – a parallel version of the same numbered items, not what respondents read. Only the instrument’s endpoints are labelled (‘strongly disagree’, ‘strongly agree’); the paper glosses them as ‘very much disagree’ / ‘very much agree’. The deposit’s scoring formula confirms which items are reverse-keyed, but order within the reverse block, and SCC_6 versus SCC_11, rest on the instrument’s numbering. Responses are stored raw: the ten reverse-keyed items (all but SCC_6 and SCC_11) are not yet reversed.
qi_2025_self_construal	The Self-Construal Scale was administered to Chinese university students, but no Chinese wording is published in the OSF deposit or the paper, so the item text is Singelis’s English original, a parallel version rather than what respondents read. The study’s own anchors (1 = very much disagree, 7 = very much agree) are used; the intermediate points are unlabelled. The deposit’s scoring formulas confirm which items form the independent and interdependent subscales, and they match the published scale item for item.
qi_2025_self_esteem	The Rosenberg Self-Esteem Scale was administered in Chinese, but no Chinese wording is published in the OSF deposit or supplement, so the item text is the RSES English original (University of Maryland form), not what respondents read. The study prints no item list and the RSES circulates in two orderings; the one shipped (reverse-worded items at Esteem_3, 5, 8, 9 and 10) reproduces the paper’s alpha of .89 where the other does not, but order within each polarity class is not pinned. Responses are stored raw and run 1 = strongly disagree to 4 = strongly agree as the paper states – the reverse of the numbering on the instrument’s own form, whose anchor words are used here.
qiang_2025_abusive_supervision	The item text here is the authors’ own English from the column headers of the study’s S1 File, but the questionnaire was translated into Chinese and administered to Chinese retail staff, and no Chinese wording for these items is published, so these are not the sentences respondents read. The paper never names or analyses this four-item block (questionnaire items 38-41), so the instrument name is IRW’s description; the items resemble four of Tepper’s (2000) abusive-supervision items but are reworded. Only the endpoints of the 1-5 scale are labelled.
qiang_2025_coping	The item text here is the authors’ own English from the column headers of the study’s S1 File, but the questionnaire was translated into Chinese and administered to 396 Chinese retail staff, and no Chinese wording for these items is published, so these are not the sentences respondents read. The paper never names or analyses this three-item block (questionnaire items 30-32); the label “coping” was assigned by IRW from item content, not by the authors. Only the endpoints of the 1-5 scale are labelled.
qiang_2025_job_dissatisfaction	The item text here is the authors’ own English from the column headers of the study’s S1 File, but the questionnaire was translated into Chinese and administered to Chinese retail staff, and no Chinese wording for these five items is published, so these are not the sentences respondents read. Only the endpoints of the 1-5 scale are labelled.
qiang_2025_job_performance	The item text here is the authors’ own English from the column headers of the study’s S1 File, but the questionnaire was translated into Chinese and administered to Chinese retail staff, and no Chinese wording for these items is published, so these are not the sentences respondents read. The paper does not analyse or cite a source for this block (questionnaire items 17-21), so the instrument name is IRW’s description. Only the endpoints of the 1-5 scale are labelled.
qiang_2025_workplace_friendship	Item fr5 is questionnaire item 35, which the IRW processing script had dropped as a presumed Chinese duplicate of item 36. The response data shows they are different administered items – they agree on only 81 of 396 rows, r = 0.208 – so it was restored on 2026-09-10 and the response table rebuilt from 1,584 to 1,980 rows. It is coded fr5 rather than inserted in questionnaire order because fr1-fr4 were already published item codes. Item 35 is the one item in this block the deposit prints only in Chinese, so unlike the other four its English here was written by IRW rather than by the study.
qin_2025_behavioral_intention	The item text here is the authors’ own English from the paper’s S1 Table, but the 547 respondents were domestic Chinese tourists answering a Chinese-language questionnaire, and no Chinese wording is published in the article or its supplements, so these are not the sentences respondents read. The source never states the 1-6 scale’s anchors, so option_text is blank.
qin_2025_cultural_identity	The 547 respondents were domestic Chinese tourists answering a Chinese-language questionnaire, but no Chinese wording is published in the article or either of its supplements, so the item text shown here is the authors’ own English from S1 Table; the 1-6 response scale’s anchors are never stated by the source, so the response options are left blank rather than guessed
qin_2025_perceived_cultural_distance	The item text here is the authors’ own English from the paper’s S1 Table, but the 547 respondents were domestic Chinese tourists answering a Chinese-language questionnaire, and no Chinese wording is published in the article or its supplements, so these are not the sentences respondents read. The source never states the 1-6 scale’s anchors, so option_text is blank.
qin_2025_tourism_experience	The item text here is the authors’ own English from the paper’s S1 Table, but the 547 respondents were domestic Chinese tourists answering a Chinese-language questionnaire, and no Chinese wording is published in the article or its supplements, so these are not the sentences respondents read. The source never states the 1-6 scale’s anchors, so option_text is blank. RCTE4 and RCTE5 keep the source’s duplicated “gain a gain a” as printed.
qin_2025_tourist_satisfaction	The item text here is the authors’ own English from the paper’s S1 Table, but the 547 respondents were domestic Chinese tourists answering a Chinese-language questionnaire, and no Chinese wording is published in the article or its supplements, so these are not the sentences respondents read. The source never states the 1-6 scale’s anchors, so option_text is blank.
queiros_2018_qcae	Two things to know. The response options ship in the administered Portuguese (‘Discordo Fortemente’ … ‘Concordo Fortemente’) with the QCAE authors’ own English alongside, and all 31 items are stored RAW – so the four reverse-worded items (QCAE_1, QCAE_2, QCAE_17, QCAE_29) still need reversing before scoring; the source deposit’s pre-reversed copies of those four were dropped by the IRW build. Separately, the English instructions are an IRW translation of the Portuguese, because this adaptation’s instruction text was never published in English. The English item and option wording, by contrast, is the QCAE authors’ own.
quinn_2023_roommate_cesd	Items 4, 8, 12 and 16, the positively worded CES-D items, are stored raw rather than reverse scored, so the 0-3 frequency anchors apply as printed to all 20 items. The study reverse-scores those four: doing so reproduces the paper’s published T1 total, while the raw sum does not.
rahm_2017_panas	The 20 item words are the administered German adjectives from the study’s SPSS labels, with the authors’ English gloss alongside, but the response anchors are English (“very slightly or not at all” … “extremely”): neither the deposit nor the paper records the German anchor wording participants saw. No instructions ship, and the reference period administered is not stated.
rahman_2022_phq9	The study’s two sources disagree on PHQ-9 item 8: the S1 Table prints only “Moving or speaking so slowly that other people could have noticed?”, while the deposited Stata label continues “Or the opposi…” before Stata’s 80-character limit cuts it off. The shorter S1 Table wording ships, so the item text may lack the second clause respondents read.
ramadan_2026_ai_awareness	Item wording for this table is the administered Arabic with the depositor’s own English alongside it, but the 1-5 response anchors are shipped in English only (strongly disagree … strongly agree): the Mendeley deposit publishes the scale labels solely as an English coding string and never gives the Arabic anchors respondents actually read
ramadan_2026_applied_practice	The Arabic wording of the five response anchors is not published in the deposit, so option_text carries the codebook’s English anchors (5 = strongly agree to 1 = strongly disagree) while the item wording is the administered Arabic.
ramadan_2026_attitudes	The item wording ships in the administered Arabic with the authors’ own English alongside, but the five response anchors ship only in English: the deposit states the 1-5 coding in English in its codebook and never prints the Arabic anchors respondents read.
ramadan_2026_perceived_competence	Item wording for this table is the administered Arabic with the depositor’s own English alongside it, but the 1-5 response anchors are shipped in English only (strongly disagree … strongly agree): the Mendeley deposit publishes the scale labels solely as an English coding string in its codebook and never gives the Arabic anchors respondents actually read
ravenscroft_2017_transition	The parent survey was administered in English, Bulgarian, Catalan, Greek and Romanian, but only the English master questionnaire is published, so the text is what 157 of the 306 respondents read and an English equivalent of what the other 149 read. The response options are printed positive-first in the questionnaire, but the stored coding runs the other way: resp 5 is the first-printed (most positive or most frequent) option and resp 1 the last. Question text is matched to the data file’s terse column labels by questionnaire order.
rd_ppcvdos_padconacns_jinbo_2019_dos	The item text here is the administered Chinese, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. Item codes carry the source numbering inside the text (4. ...) and are left verbatim.
rdatasets_gssabortion	Item text comes from the published original instrument rather than this study’s own materials
regalado_2023_tourism_value	Item P4_14 ships with no Spanish item text: the study’s Spanish SPSS file repeats P4_15’s label on P4_14, so the administered wording is unrecoverable, and only the authors’ English (“In general, Valuable and worth it”) is given. The paper describes 12 items rated from “not important” to “very important”, but the data hold 15 items (P4_13-P4_15 are overall evaluations) and the file’s own anchors, which ship, run “Nada valioso” to “Muy valioso”.
reinecke_2018_online_vigilance	The Online Vigilance Scale was administered in German, but neither the article nor its Supporting Information datasets publish any German wording, so the item text and the two endpoint anchors here are the English of the study’s own SPSS variable and value labels, not the sentences respondents read. Scale points 2-4 carry no option text.
reinwarth_2023_domain_importance	The German item and option text is the FLZM questionnaire as respondents saw it, taken from a published reproduction of the instrument rather than the study’s terse data labels. The English in the _translated columns is Schwinn et al.’s (2026) printed translation, and its label for resp 2, ‘rather not important’, is a loose rendering of the administered German ‘Etwas wichtig’ (somewhat important).
ren_2024_loneliness	Respondents answered a Chinese-language version; the item text shown is the study’s own short English rendering from its data file, and only the end anchors were published by the study, so ‘Some of the time’ is the standard wording of the three-item scale.
ren_2024_phq9	Respondents answered a Chinese-language PHQ-9 whose wording the study did not publish; the item text is the standard English PHQ-9, and resp 1-4 are the standard 0-3 scoring shifted up by one (1 = Not at all, 4 = Nearly every day). The data carry no item labels, so which code is which item rests on the standard PHQ-9 numbering; item 9 and the pair of items 1-2 are confirmed from the data, the order of items 3-8 is not.
Resistance	resp is not a rating: it is the absolute difference (0-5) between the respondent’s ratings of the two frames of each framing problem, so each item_text carries both administered wordings under ‘Gain frame:’/‘Loss frame:’ or ‘Positive frame:’/‘Negative frame:’ labels that IRW added, and option_text is blank. The wording is the English version, which only the 261 North American respondents read; the other 245 answered a validated Bulgarian translation that is not shipped.
reyes_2022_eheals	The study’s survey and data file do not reproduce the eHEALS wording – both point to Table 1 of Norman & Skinner (2006) – so the item text and instructions are the published instrument’s. The instructions keep the interviewer phrasing “tell me which response best reflects your opinion” although this study fielded the scale online. The response anchors follow the study’s own data key, 1 = Strongly agree to 5 = Strongly disagree, which is the reverse of the instrument’s own coding, and the midpoint reads ‘Neither agree nor disagree’ where the instrument prints ‘Undecided’.
rfq_wozniakprus_2021	The RFQ-8 was administered in Polish, but neither the deposit nor the closed-access paper contains Polish item wording and the rights holder’s official Polish form is no longer distributed, so the item text is the canonical English RFQ-8, not what respondents read. All eight items run 1 = Strongly disagree to 7 = Strongly agree; the source file labels RFQ8 in the opposite direction, which is an error in the source and is overridden here. In IRW versions before the 2026-09 correction, RFQ8’s response values 1 and 7 were swapped (88 of 538 responses); results that scored RFQ8 from those versions should be rerun.
rfq8_wozniakprus_2022	The RFQ-8 was administered in Polish, but neither the study’s deposit nor its paper contains the Polish wording, so the item text is the canonical English RFQ-8, copied from another study’s published table, standing in for what respondents read. All eight items run 1 = strongly disagree to 7 = strongly agree, with only the endpoints labelled; the source SPSS file labels RFQ8’s scale the other way round, which the study’s own scoring columns show to be a labelling error, so it is overridden here. RFQ7 is reverse-keyed and stored unreversed.
riasec	Only items 1-48 of this table are the RIASEC interest inventory the table is named for. Items 49-58 are the Ten Item Personality Inventory and items 59-74 a 16-word vocabulary checklist, administered in the same session, and the three use different response scales (1-5, 1-7 and 0/1 respectively)
ribeiro_2019_academic_motivation	The English item and anchor wording for this table was translated by IRW from the study’s Portuguese; the source publishes no English version of the scale. The seven response labels come from the deposit’s SPSS value labels, which name every point. The paper’s Annex prints only five labels across the seven points, with ‘Moderada correspondencia’ spanning 3-5 and ‘Pouca correspondencia’ at 2 where the data file has ‘Muito Pouca Correspondencia’.
rinaldi_2021_mas	This table’s 61 items were administered in Italian, but no Italian wording is published in the study’s paper or in its only supplementary file (an SPSS data file that carries no variable or value labels for the items), so IRW ships the original English MAS wording that the paper reproduces in its own Table 2. Separately, that table documents 60 items for 61 data columns: the alignment verified against the published per-item means, the reverse-keying pattern and the paper’s factor structure is Mas_1-Mas_56 to Table 2 items 1-56 and Mas_58-Mas_61 to items 57-60, leaving Mas_57 with no published wording at all – its item_text is blank rather than guessed
risticdedic_2025_dhq_currentstate	The DHQ was administered in ten languages across nine EU countries (Croatian, Dutch, Finnish, German, English, Catalan, Slovenian, French, Greek and Spanish); the item text shipped here is the English master version, which 87 of the 834 schools answered directly and from which the other nine versions were translated. All ten language versions are published by the authors at doi.org/10.5281/zenodo.13767190. The 0-100 slider is labelled only at its endpoints, so option_text is populated for resp=0 and resp=100 only.
risticdedic_2025_dhq_expectation	The DHQ was administered in ten languages across nine EU countries (Croatian, Dutch, Finnish, German, English, Catalan, Slovenian, French, Greek and Spanish); the item text shipped here is the English master version, which 87 of the 834 schools answered directly and from which the other nine versions were translated. All ten language versions are published by the authors at doi.org/10.5281/zenodo.13767190. The 0-100 slider is labelled only at its endpoints, so option_text is populated for resp=0 and resp=100 only.
risticdedic_2025_dhq_importance	The DHQ was administered in ten languages across nine EU countries (Croatian, Dutch, Finnish, German, English, Catalan, Slovenian, French, Greek and Spanish); the item text shipped here is the English master version, which 87 of the 834 schools answered directly and from which the other nine versions were translated. All ten language versions are published by the authors at doi.org/10.5281/zenodo.13767190. The 0-100 slider is labelled only at its endpoints, so option_text is populated for resp=0 and resp=100 only
ritzel_2020_farmer_burden	The German wording comes from the study team’s companion report of the same survey, not from the deposit. Items y6, y7 and y8 are shown as printed in German, positively phrased (e.g. ‘Ich identifiziere mich…’), but their stored responses follow the paper’s negated English (high = does not identify / not important / not appropriate), so for these three items resp 1 is labelled ‘sehr stark’ and resp 7 ‘überhaupt nicht’. The English for y6-y8, x1’s question and several middle response categories was produced by IRW; the rest is the paper’s own English. x1 is stored on the questionnaire’s eight categories, while the paper collapses it to six. Whether French or Italian versions were also fielded is not documented.
rodriguezandres_2016_mnemocity_usability	The item wording here was read from the article’s Table 2 and Table 3, which PLOS publishes only as images, so it is worth a spot-check; two caveats on content. First, item SA1 asks ‘How much fun did you have?’ but Table 2 gives it an agreement scale (Strongly disagree..Strongly agree) while Table 3 gives the identical question a Very boring..Very fun scale – the source’s Table 2 anchors are shipped as printed rather than corrected, so SA1’s response options may be a copy-paste error in the paper. Second, the 160 respondents were Spanish children aged 5-10 who answered in Spanish, but no Spanish wording appears in the article or its single supplement, so the English shown is the authors’ own from those tables.
rodriguezmuniz_2016_washback_survey	Administered to secondary-school mathematics teachers in Asturias, Spain, presumably in Spanish, but the Spanish wording is not published, so IRW ships the English the paper prints in its Table 8. Item means reproduce the paper for all 17 items, but Q11 and Q16 are too close to tell apart from the data (printed means 3.13 and 3.12), so their assignment rests on the paper’s numbering.
rodriguezquiroga_2024_epistemic_trust	The Spanish item wording is the study’s own; the English alongside it is Campbell et al. (2021)’s official ETMCQ, while the English instructions and the four intermediate response-scale labels were written by IRW because the instrument’s publisher printed only the 1/4/7 anchors.
roelen_2020_dietary_diversity	This household food-group checklist was administered in Haitian Creole and the study’s deposit publishes no Creole wording, so the item text shown is the English of the dataset’s own Stata variable labels; two of those labels hit Stata’s 80-character cap and are therefore shipped truncated in mid-phrase (q_306 ends ‘Irish potatoes, brea’ and q_311 ends ‘coconut, lemon,’), and the deposit publishes no framing question for the block, so no instructions or section prompt are given
roelen_2020_k6	The K6 here was administered in Haitian Creole but only English wording was ever published, so the item text shipped is the study’s English rendering from its Stata deposit and the paper’s Table 3, not what respondents heard; note also that the IRW resp coding runs 0 = ‘None (0 day)’ to 4 = ‘All (30 days)’ (higher = more distress), which is the reverse of the direction the paper’s own K6 total score uses (its mean of 19.0 is a sum of 1 = all 30 days .. 5 = none).
roelen_2020_rses	This table’s 10 Rosenberg Self-Esteem Scale items were administered orally in Haitian Creole to women in rural Haiti, but no Creole wording is published in the paper or in its single supplementary data file, so IRW ships the English wording that the study’s own Stata file and its Table 3 carry. Note also that responses are stored RAW: the five negatively worded items (q_408, q_411, q_412, q_414, q_415) are not reverse-scored in this table, and the deposited Stata file’s own numeric codes run in the opposite direction to the IRW resp values – the anchors shipped here (1 = Strongly disagree … 4 = Strongly agree) were confirmed against the data, with all 40 item-by-level counts matching and q_414 reproducing the paper’s reported 94.7% agree/strongly-agree exactly
roettl_2018_brand_recognition	resp in this table is the study’s 0/1 scoring of a four-alternative forced-choice logo recognition item, not the choice itself, so option_text carries the scoring labels (‘wrong’/‘right’) rather than the four logo names respondents actually saw; those alternatives are listed per category in the paper’s S2 Appendix.
roettl_2018_scepticism	The nine items are the SKEP advertising-scepticism scale’s pro-advertising statements (‘Advertising is generally truthful’), stored raw and agreement-keyed 1-7 with no reverse scoring applied, so despite the table’s name a HIGH resp means LOW scepticism. Live per-item means run 2.15 to 3.50 (n=234 on all nine). This is a property of the response data, not of the item text. Two smaller notes: the study’s S2 Appendix numbers this block F29_1-F29_9 while the data columns, and so the IRW item codes, are F31_1-F31_9 – the two were tied by exact string identity of the English wording, not by position; and only the two scale poles carry labels in the source, so option_text is blank for resp 2-6 rather than padded.
rogers_2021_financial_knowledge	The response values are right/wrong scores rather than the option each student chose, so the response options shown here are the data file’s own ‘Resposta incorreta’/‘Resposta correta’ labels; the questionnaire’s three-to-five multiple-choice alternatives (including a ‘Nao sei’ option) are published in the deposit’s FinalForm.pdf but cannot be attached to a 0/1 correctness code, and the English translations beside the Portuguese wording were produced by IRW, not by the study’s authors
rogowska_2023_maia2	Items MAIA2_05 to MAIA2_12 and MAIA2_15, the reverse-keyed Not-Distracting and Not-Worrying items, have no response-option labels. The questionnaire labels 0 “Nigdy” (never) and 5 “Zawsze” (always), but the deposit does not say whether these nine items were stored as answered or already reverse-scored, and the evidence in the data points both ways, so a stored 0 may mean either. The other 28 items carry both anchors. The English is the canonical MAIA-2 (Mehling et al. 2018), not a back-translation of the Polish.
rosenberg_fadplus_goto2021	The item text here is the administered Japanese, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. Only option_text is translated: item_text is already English in the published table, for a Japanese administration.
rosetti_2023_cas	The Rosetti et al. study was run with Mexican university students and was almost certainly administered in Spanish, but neither the figshare deposit (a single fully anglicised CSV) nor the closed-access Ecopsychology article publishes the Spanish wording, so the item and option text shipped here is Clayton & Karazsia’s English original rather than the sentences respondents actually read.
rosetti_2023_gad7	The GAD-7 was administered in Spanish to Mexican university students, but no Spanish wording is published in the study’s figshare deposit and its paper is closed access, so the item and option text shipped here is the instrument publisher’s official English (phqscreeners.com) rather than the sentences respondents read
rosetti2023_gad7	The GAD-7 here was administered to Mexican university students, but the figshare deposit publishes no item wording and the source paper is closed access, so the item and option text shipped is the instrument publisher’s official English (phqscreeners.com) rather than the Spanish sentences respondents read; the mapping of GAD7_q1..q7 to the canonical GAD-7 items rests on the column numbering, and only q5 (restlessness) and q6 (irritability) are independently pinned by the data
rosharudin_2023_dass21	The live table records five response levels (0-4), but the paper and the Malay DASS-21 form it administered both describe a four-point 0-3 scale with four labelled anchors, so no option text and no instructions could be shipped for this table; the fifth level is genuine rather than a missing-value code, since counting it as 4 reproduces the paper’s published subscale standard deviations to within 0.02.
rosharudin_2023_ders18	The three Awareness items (@1_A, @4_A_A, @6_A) are stored RAW in the IRW table, so a higher response means MORE emotional awareness – the opposite direction from the other 15 items, and the reverse of how the source paper’s Table 2 reports them. Reverse-code them before summing a DERS-18 total. Verified from the data: each of the three matches its published floor/ceiling percentages only with floor and ceiling swapped (e.g. item 6, published 11.47/18.72 vs observed 18.72/11.47). The five response-option labels are the canonical English DERS-18 anchors, not the Malay wording respondents saw, which the study never published.
rothlin_2022_scs_sf	The item text is the English SCS-SF wording carried in the deposit’s variable labels; the questionnaire was given in German, and the German wording respondents read is not published.
roy_2024_phq9	The paper does not state the administration language. It cites a Bengali adaptation of the PHQ-9, so the questionnaire was probably administered in Bengali, but no Bengali wording is published, so the item and option text is the publisher’s official English (phqscreeners.com) rather than necessarily the sentences respondents read. Item order is assumed to follow the standard PHQ-9 numbering.
roy_2024_sas_sv	The paper does not state the administration language. The questionnaire was probably administered in Bengali, but no Bengali wording is published, so the item and option text is the scale developers’ official English (Kwon et al. 2013, Appendix S1) rather than necessarily the sentences respondents read. Item order is assumed to follow the published numbering.
rr98_accuracy	Item text for this table describes the visual stimulus, not a question read to the participant: rr98’s items are pixel arrays, so IRW composed each item_text from the source package’s own definition of the stimulus-strength variable rather than transcribing administered wording, which does not exist. Note also that resp is trial accuracy (1 = the response matched the distribution the stimulus was drawn from), not the participant’s dark/light choice.
ruiz_parra_2023_rfq8	The RFQ-8 was administered here in Spanish. The English item and anchor wording is the original English RFQ printed in the study’s S1 Appendix – a parallel published version rather than a translation of the Spanish items – but the English instructions (instructions_translated) were translated by IRW from the Spanish and are not the study’s own words. Item 7 (“Siempre se lo que siento”) is reverse-keyed and is stored unreversed. The instrument’s authors note that the RFQ is freely available for research but is not yet suited for clinical purposes.
RvKDCS_Romiacg_Miroshnik_2020_CBI	The English in the *_translated columns was produced by IRW from the administered Russian; the Russian items are the study’s own adaptation of Hocevar’s Creative Behavior Inventory, so this English is a rendering of what respondents read rather than Hocevar’s original wording.
rvobgvmaas_lsf_lehing_2024	Items are shown in the administered German (Wittich et al. 2009), with Condon’s original English alongside; the English is the source instrument, not a translation of the German, and the two differ in places (the German instructions add ‘Es gibt keine richtigen oder falschen Antworten.’). In the response table, the wave-1 rows for the 11 reverse-keyed items (codes ending r_t1) are copies of the wave-0 rows (2,024 of 2,024 identical) because of an IRW processing-script error, so the study’s wave-2 answers to those items are not in IRW.
rzeszutek_2020_wwii_matgrandfather	Each of the 29 questions was administered with three responses - TAK (yes), NIE (no) and NIE WIEM (don’t know) - but the IRW table drops ‘don’t know’ as missing, so only the two option rows 0=NIE and 1=TAK exist and per-item n runs 84-263 of 500 respondents; the item wording is the questionnaire’s shared grandmother/grandfather row text, and this table is specifically the mother’s-side (great)grandfather column of the grid
rzeszutek_2020_wwii_matgrandmother	Each of the 29 questions was administered with three responses - TAK (yes), NIE (no) and NIE WIEM (don’t know) - but the IRW table drops ‘don’t know’ as missing, so only the two option rows 0=NIE and 1=TAK exist and per-item n runs 106-322 of 500 respondents; the item wording is the questionnaire’s shared grandmother/grandfather row text, and this table is specifically the mother’s-side (great)grandmother column of the grid
rzeszutek_2020_wwii_patgrandfather	Each of the 29 questions was administered with three responses - TAK (yes), NIE (no) and NIE WIEM (don’t know) - but the IRW table drops ‘don’t know’ as missing, so only the two option rows 0=NIE and 1=TAK exist and per-item n runs 76-216 of 500 respondents; the item wording is the questionnaire’s shared grandmother/grandfather row text, and this table is specifically the father’s-side (great)grandfather column of the grid
rzeszutek_2020_wwii_patgrandmother	Each of the 29 questions was administered with three responses, TAK (yes), NIE (no) and NIE WIEM (don’t know), but the IRW table drops ‘don’t know’ as missing, so only the two option rows 0=NIE and 1=TAK exist and per-item n runs 90-238 of 500 respondents. The item wording is the questionnaire’s shared grandmother/grandfather row text, and this table is the father’s-side (great)grandmother column of the grid. Item owojb23 is grouped with items 9-22 as in the English form, although the Polish father’s-side form prints it under the ‘witness to’ heading.
sabers_vesper2023	The four reverse-worded items (negaf1_r, negaf2_r, negaf3_r, leg3_r) are stored reverse-scored, so their option text runs from Agree (1) to Do not agree (5). Item text is the English version; 454 German and 463 French respondents answered the authors’ translations, which are published in the OSF deposit but not shipped here.
SABFI2_Gallardo_Pujol_2018_Constru	The item codes Constru1-Constru13 are matched to wording by the order the International Situations Project survey displayed the items, which the data confirm by subscale and by which pole of each subscale an item belongs to but not within a pole; the English beside Constru1 and Constru11 is the survey’s English original, and the administered Spanish differs from it (Constru1 asks about saying what you think even when it is inappropriate for the situation, Constru11 about seeing yourself differently in different social environments).
SABFI2_Gallardo_Pujol_2018_LOT	This study administered the Spanish LOT-R, whose wording is not in the study’s deposit or paper, so item_text is the scale authors’ own English LOT-R standing in for it and the response labels are the paper’s English description of a five-point strongly disagree to strongly agree scale; LOT2, LOT4 and LOT5 are the pessimism items and are stored unreversed, and which optimism (or pessimism) item each code denotes follows the published item order, which the data confirm by polarity class but cannot confirm within a class.
SABFI2_Gallardo_Pujol_2018_Micro	The item codes Micro1-Micro6 are not a racial microaggression scale but the six contexts of the micro-macro life-space question (Salgado et al., 2013) asked in the International Situations Project; which of family, friends and work each of Micro1, Micro2 and Micro6 denotes follows the survey’s screen order, which the data confirm for the city/society/world block but cannot confirm within the family/friends/work block.
SABFI2_Gallardo_Pujol_2018_Tight	The item codes Tight1-Tight6 are matched to wording by the order the International Situations Project survey displayed the items; the data confirm that Tight4 is the reverse-worded ‘freedom’ item but cannot confirm the order of the other five, and item 4’s Spanish reproduces the survey’s own typo ‘les personas’.
SABFI2_Gallardo_Pujol_2018_Trust	The item codes Trust1-Trust5 are matched to wording by the order the International Situations Project survey displayed the items; the data support placing the first-person item (‘I trust others’) at Trust4 and ‘Most people trust others’ at Trust3 but cannot separate Trust1 from Trust2, and item 5’s Spanish reproduces the survey’s own typo ‘depersonas’.
safiye_2023_rfq	The RFQ-8 was administered to this sample in Serbian, but neither the study’s data deposit nor its paper contains the Serbian item wording, so the item text shipped here is the standard English RFQ-8 wording standing in for what respondents actually read. The response labels are the study’s own English labels from its data file. Item RFQ7 (“I always know what I feel”) is reverse-keyed and is stored unreversed. The instrument’s authors note that the RFQ is not yet suited for clinical purposes.
saha_2026_cesd	The wording is this study’s question-form rephrasing of the CES-D (e.g. ‘Did you feel depressed?’), not the original statements. The positively worded items Q4, Q8, Q12 and Q16 are stored reverse-scored, so for them resp 0 is ‘Most or all of the time’ and 3 is ‘Rarely or none of the time’.
salleh_2023_aim_iam_fim	The item wording is the data file’s combined “Program MIICA/ KOSPEN” label. These 170 respondents are the study’s KOSPEN pilot, but the only questionnaire the study published is the MIICA form, which names a single programme – so the programme name these respondents actually read is not confirmed. Seven item labels were truncated by one character in the data file and have been restored from the paper’s Table 2.
sapa_personality	17 of the 696 items ship the SAPA item dictionary’s abbreviated shorthand – bracketed or elided text – rather than the full sentence the participant read; the remaining 679 carry the full wording. Item text and the six response anchors both come from the SAPA-Project data release’s own item dictionary (ItemInfo696) and code book
SBD_Smith_2020	The response table omits one of the study’s 303 respondents (the first data row was read as a header by the IRW processing script), so it holds 302.
schafer_2016_music_effects	The diary was administered in German and no German wording is published, so the item text is the author’s own English rendering from the paper’s Method section. The study’s data file labels the third item “emotion and mood” while the paper’s question words it “arousal and mood regulation”; that item’s mean, 6.17, matches the paper’s 6.2 for the same function.
schafer_2016_music_goals	The diary study was run at a German university and was very likely administered in German, but neither the paper nor its data supplements publish the German questions, so the item text here is the English rendering printed in the author’s own article. Scale points 2-9 carry no labels in the source and are left blank rather than padded.
schalet_2016_hrsa	The study published no item wording for its anxiety measure and its data file carries only generic variable labels, so the item and option wording here is the canonical Hamilton Anxiety Rating Scale taken from a public-domain reproduction sheet. The instructions shown are that sheet’s rater instruction and have not been confirmed as the ones this trial used. Subscale membership is pinned by the paper’s Table 2 and 10 of 14 items are pinned individually, but the order within two pairs (a03/a05 and a07/a08) is not verified.
schmidt_2017_fas	The item and response text is the German the study administered, taken from the deposited SPSS file’s own variable and value labels (including a lowercase first letter on the bedroom item and a missing question mark on the holidays item); no printed questionnaire is published, so these labels are the only record of the wording. It is the study’s own German rendering of the HBSC Family Affluence Scale II, not text from the HBSC protocol. The English in the _translated columns was produced by IRW, not by the study’s authors.
schmidt_2017_pds	Items pub_2_1 and pub_3_1 each hold two sex-specific questions under one code (girls: breast growth and menstruation; boys: beard growth and voice change), and for pub_3_1 the lowest and highest levels combine the girls’ yes/no answers with the boys’ anchors. The item text is the study’s own abbreviated German data-file labels, ending in ellipses, rather than the full administered questionnaire wording, and the English in the _translated columns was produced by IRW rather than by the study’s authors.
science_ltm	For Environment, Technology and Industry – the three negatively worded statements – resp is reverse-scored relative to the ltm package’s documentation and to this table’s own resp_raw column: resp 4 means the respondent strongly DISAGREED with the statement as worded. The Eurobarometer 38.1 Great Britain codebook marginals confirm the direction shipped here (Industry: 39% strongly disagree in the survey against 41% at resp 4 in this table), so option_text follows the data rather than resp_raw.
scs_2025_cfaclinic	The scale was administered in Turkish, but the study published Turkish wording for only six of the 12 items (1-5 and 11), which carry it with the paper’s English alongside. The other six items, the instructions and the response anchors are Neff’s official English SCS-SF, not what respondents read. Items 1, 4, 8, 9, 11 and 12 are stored reverse-scored, so on those items resp 1 corresponds to ‘Almost always’ and resp 5 to ‘Almost never’.
scs_2025_cfanonclinic	The scale was administered in Turkish, but the study published Turkish wording for only six of the 12 items (1-5 and 11), which carry it with the paper’s English alongside. The other six items, the instructions and the response anchors are Neff’s official English SCS-SF, not what respondents read. Items 1, 4, 8, 9, 11 and 12 are stored reverse-scored, so on those items resp 1 corresponds to ‘Almost always’ and resp 5 to ‘Almost never’.
scs_2025_cfausatr	The English SCS-SF wording shown is what the 125 US respondents read; the 545 Turkish respondents answered the authors’ Turkish adaptation, which was not published in full. The instructions are Neff’s standard sentence, as the study did not publish its own. Items 1, 4, 8, 9, 11 and 12 (the negatively worded items) are stored already reverse-scored, so for those items resp 1 = Almost always and 5 = Almost never.
selfcompassionscale_shortform_fuochi_2025	The item text is the administered Italian, transcribed from the paper’s Table 2. The English in the _translated columns is Neff’s official English SCS-SF, a parallel version rather than a translation of the Italian, and the English instructions are Neff’s instruction block, not a rendering of the Italian one. The study’s Italian short-form wording also differs slightly from the Italian 26-item SCS it was drawn from (items 1, 8, 9 and 10).
sem_cnes	The 1997 Canadian Election Study mail-back questionnaire was administered in English or French; this table carries the English statements only, as documented in the R sem package. The response labels (Strongly disagree to Strongly agree) are rendered from that package’s factor levels, not transcribed from the printed questionnaire, which could not be retrieved.
shen_2025_bsmas	As for shen_2025_dass21: the supplement’s English wording, not the Chinese wording respondents read.
shen_2025_dass21	The item text and response options come from the English SPSS labels and codebook in the article’s supplementary files, the authors’ rendering of a questionnaire administered in Chinese to language teachers in China; the Chinese wording respondents read is not published.
shen_2025_sabas	As for shen_2025_dass21: the supplement’s English wording, not the Chinese wording respondents read.
shi_2021_gentrification	The questionnaire was distributed mainly through a Chinese survey platform to respondents who were almost all Chinese nationals, but the study publishes only English factor names, so the item text shown is the authors’ English and may not be the wording most respondents read. The paper does not state the administered language.
shi_2025_dass21	Response codes are not consistent across waves in this table. Baseline (wave 0) responses run 1-4, while post-intervention and follow-up (waves 1-2) run 0-4 with five levels – against the four-point 0-3 scale the source paper describes. Because no single labelling is correct across all three waves, the response options are left unlabelled. The item text is the canonical English DASS-21, as the administered simplified Chinese wording was not published.
shin2024_creactability_adaptability	The scale was administered in Korean (coaches rating their players), but no Korean wording is published, so the item and response text is the English rendering printed in the authors’ article rather than what raters read. The article lists the three adaptability items without numbers, so which wording belongs to Adaptability1-3 follows the order listed; the data confirm that order matches the paper’s item numbering but cannot confirm it for the wording itself.
shin2024_creactability_creativity	The scale was administered in Korean (coaches rating their players), but no Korean wording is published, so the item and response text is the English rendering printed in the authors’ article rather than what raters read. The article lists the three creativity items without numbers, so which wording belongs to Creativity1-3 follows the order listed; the data confirm that order matches the paper’s item numbering but cannot confirm it for the wording itself. The ‘not at all’ label for response 1 is absent because no rating used it.
shin2024_creactability_quickness	The scale was administered in Korean (coaches rating their players), but no Korean wording is published, so the item and response text is the English rendering printed in the authors’ article rather than what raters read. The article lists the three quickness items without numbers, so which wording belongs to Quickness1-3 follows the order listed; the data confirm that order matches the paper’s item numbering but cannot confirm it for the wording itself.
shineha_2024_gef_accept_factors	This survey was administered in Japanese and the Japanese wording is not in the study’s supplements, so item_text is the authors’ own English translation of the questionnaire, which they describe as tentative. Each item is a 0/1 indicator of whether the respondent chose that factor as one of three important for social acceptance; respondents who chose “There are no important things” chose only that option.
shineha_2024_gef_attitudes	The survey was administered in Japanese and only the authors’ English translation of the questionnaire is published, so the wording shown is that translation. For Fig1b (“accepted in society”) the translated questionnaire prints Strongly agree…Strongly disagree anchors, which are the ones shipped here, while the study’s data file labels the same codes Strongly acceptable…Strongly unacceptable. Both run in the same direction, with 7 the most positive. The item codes are not questionnaire order.
shineha_2024_gef_info_topics	The questionnaire and the paper both instruct respondents to “choose three” topics, but in the data every respondent selected either exactly five topics (3,614) or only “Nothing in particular that I need to know” (784) – and the paper’s own Figure 3 percentages, which sum to well over 100%, were computed from these same five-selection data. The public and expert samples also saw different prompts (“What do you want to know?” against “What do you want to inform?”). The English throughout is the authors’ translation of a Japanese questionnaire, and option_text is blank because the 0/1 selection indicators carry no printed labels.
shinohara_2021_testimony	Item codes are puppet roles, not fixed puppets: which coloured puppet and which video version filled each role, and the presentation order, were counterbalanced, and the valence-testimony and first-hand-observation puppets were prosocial for the positive-condition half of the sample and antisocial for the negative-condition half (cov_condition). Allocation and anticipation are stored as ranks where 1 is the highest reward or first choice and 3 the lowest. The English prompts are the paper’s descriptions of a procedure administered in Japanese, not a published script
short_dark_triad	The item codes are bare integers 1-27 assigned by IRW processing; the source’s own codes, which name the subscale each item belongs to, are not carried in the table. In code order, items 1-9 are Machiavellianism, 10-18 Narcissism and 19-27 Psychopathy. The deposit states only three of the five scale anchors (1 = disagree, 3 = neutral, 5 = agree), so option_text is blank at 2 and 4 rather than padded, and it records no participant-facing instructions, so instructions is empty.
shorttripm_silvia_2025	The four items whose codes end in r (tripm_mean_1r, tripm_mean_13r, tripm_bold_11r, tripm_bold_14r) are stored already reverse-scored, so for them resp 1 = True and resp 4 = False, the opposite of the other items and of the coding in the study’s survey file.
shu_2024_gad7	The GAD-7 was administered in Chinese, but the study’s deposit carries only its own English rendering of the items and response options, which is what is shipped here; it is neither the Chinese respondents read nor the official English GAD-7 wording
shu_2024_phq9	This PHQ-9 was administered in Chinese to adolescents in China, but the study’s data file publishes only an English rendering, so item_text is the study’s own English (not the canonical PHQ-9 wording and not what respondents read). PHQ1’s label was truncated in the file and its final words ‘doing things’ were completed by IRW, and the instructions’ embedded anchor legend words the middle options differently from the response labels
shu_2025_translation_add	Two of the three questions are matched to their item codes on content; the third, addcompetitioninreha, is assigned by elimination and its code does not obviously describe the question it carries (pressure of current studies on career plans), so treat that wording as provisional. The questionnaire was administered in Mandarin but only the study’s English ships. No response options are published, and that item uses four levels where the other two use five.
shu_2025_translation_mcpis	The item text here is the administered Chinese, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. option_text is already English, so only item_text carries a translation.
silva_2018_mbds	resp is not a chosen response option: each MBDS item was rated for importance (1-10) and on a 5-point agreement/frequency scale, and resp is importance/10 multiplied by that rating (0.1-5.0), so option_text is blank for all 28 observed values. The Portuguese wording is the reduced-form reprint in Da Silva et al. (2017), which the study cites as the version used, not the study’s own materials
silva_2022_crsy	Item text is the English wording in the authors’ own data file; the survey of Portuguese students was presumably administered in Portuguese, and that version was not published.
silva_2023_ecohis	The study describes its ECOHIS as a 5-point scale and prints no anchor labels, but the published ECOHIS has six options including “Don’t know”, and in these data code 5 behaves like a non-ordinal answer (respondents choosing 5 have lower caries counts and lower scores on the other items than those choosing 4), so option_text is left blank and resp 5 should not be assumed to mean “Very often”. Item wording is the authors’ English rendering of the Brazilian Portuguese version they administered
silvia_2024_funny	The paper describes a 1-5 scale; this table stores it as 0-4 (the authors’ own analysis script adds 1), so the endpoint labels sit at 0 = strongly disagree and 4 = strongly agree. Points 1-3 were never labelled.
simsalRbim_Human_LargeValence_2017	Item text is blank by design because the stimuli are OASIS photographs (the item code names the image, e.g. Lake = Lake_9.jpg); each pairwise trial contributes two rows, resp 1 for the image chosen and 0 for the image not chosen, and the instructions are transcribed from the study’s publicly hosted online version of the experiment rather than a separately published copy of the 2017 laboratory script
simsalRbim_Human_LargeValence_2018	The items are OASIS photographs shown two at a time, not questions: the bracketed item_text identifying each image is a description written by IRW, not anything shown to subjects. resp is 1 when this image was chosen on that trial and 0 when the other image of the pair was.
simsalRbim_Human_LowValence_2017	Pictures, not text, were the stimuli: item codes name the OASIS image shown (Sunset_1, Flowers_3, Food_4, Grass_1, House_1, Monkey_2, Timber_1), and resp = 1 means that picture was the one chosen from its pair. The deposit contains 63 participants where the source paper reports 48 for this picture set
simsalRbim_Mice_LowValence	The items are liquids offered to mice two at a time, not questions: the bracketed item_text is a description written by IRW from the study’s methods. resp is the number of corner visits with licks for that liquid in that pairing, a count rather than a rating. The code Sac stands for sucrose, not saccharin.
sinche_2017_skill_importance	The study’s data file labels this scale strongly disagree..strongly agree, but the paper reports it was rated from “not at all important” (1) to “extremely important” (5); the paper’s anchors are shipped and the wording of points 2-4 was not published
singh_2025_identity_pba	Table has 6 items for affiliation but the codebook only has 5 items
singleton_2021_carbon_reduction	The seven even-numbered items (Q2, Q4, Q6, Q8, Q10, Q12, Q14) are the NEP’s anti-ecological statements and are stored already reverse-scored as the study scored them, so for those items resp=1 is ‘strongly agree’ and resp=5 is ‘strongly disagree’ (the option_text rows are mapped accordingly); for odd-numbered items resp=5 is ‘strongly agree’, so higher resp means a more pro-ecological response on every item
sison_2022_hiv_testing_stigma	The item text is the English wording from the deposit’s coding manual. The paper states that the survey questions “were in English and Tagalog (local language)”, and the manual codes consent as “Yes/Oo”, so respondents saw Tagalog wording, alongside or instead of the English, that is not published anywhere in the deposit or the article. The code-to-text mapping is the manual’s own and is exact; only the Tagalog is missing.
sison_2022_provider_mistrust	As for sison_2022_hiv_testing_stigma: the English wording from the deposit’s coding manual, for a survey whose questions were in English and Tagalog; the Tagalog wording respondents saw is not published.
skalka2025_ai_literacy	resp 0 in this table is a no-answer code rather than a scale point – the deposit documents only a 5-point 1-5 agreement scale, 0 occurs in just the 19 IM/S/C/BI items and is concentrated in students reporting no AI coursework, so it ships with blank option_text and should be treated as missing, not as a value below ‘strongly disagree’
skarzauskiene_2026_attitudes_science	The survey was run face-to-face in Lithuania, but the deposit publishes only the authors’ English rendering of the questions (printed in capitals) and answer options, so the text shown is that English rather than what respondents heard. Option 4 reads ‘I agree more quickly’, a literal rendering that means ‘tend to agree’, while the question’s own prompt says ‘tend to agree’. Answer code 9 (N/A) is treated as missing and is not in the response data.
skarzauskiene_2026_fake_news_agree	The survey was administered face-to-face in Lithuanian, but only the authors’ English documentation is published, so the item and option text is their English rendering rather than what respondents heard; option 4 reads ‘I agree more quickly’, a literal rendering that means ‘tend to agree’. Answer code 9 (N/A) is treated as missing and is not in the response data.
skarzauskiene_2026_fake_news_frequency	Respondents were interviewed face-to-face in Lithuanian; the item and option text is the authors’ own English rendering from the data dictionary, as the Lithuanian wording was not published. Item m51 (‘I easily recognize fake news.’) sits under the prompt ‘HOW OFTEN WOULD YOU DO THIS WHEN ENCOUNTERED WITH FAKE NEWS ON SOCIAL MEDIA?’ in that English, as printed. Answer code 9 (N/A) is treated as missing.
skarzauskiene_2026_information_sources	Respondents were interviewed face-to-face in Lithuanian; the question and answer text is the authors’ own English rendering (printed in capitals) from the data dictionary, as the Lithuanian wording was not published. Answer code 9 (N/A) is treated as missing.
skarzauskiene_2026_science_behaviors	Respondents were interviewed face-to-face in Lithuanian; the question and answer text is the authors’ own English rendering (printed in capitals) from the data dictionary, as the Lithuanian wording was not published. Answer code 9 (N/A) is treated as missing.
skarzauskiene_2026_science_engagement	Respondents were interviewed face-to-face in Lithuanian; the item and option text is the authors’ own English rendering from the data dictionary, as the Lithuanian wording was not published, and options 1 and 2 (‘It doesn’t matter at all’, ‘Rather it doesn’t matter’) keep that rendering’s literal phrasing. The table name is broader than its content: the four items are reasons for wanting to learn more about science news (‘WHY DO YOU WANT TO LEARN MORE ABOUT SCIENCE NEWS?’), not general engagement with science. Answer code 9 (N/A) is treated as missing.
skarzauskiene_2026_social_trust	Respondents were interviewed face-to-face in Lithuanian; the statement and answer text is the authors’ own English rendering from the data dictionary, as the Lithuanian wording was not published, and option 4 reads ‘I agree more quickly’, a literal rendering that means ‘tend to agree’. m71 and m72 are negatively worded trust statements and are stored unreversed. Answer code 9 (N/A) is treated as missing.
skarzauskiene_2026_trust_science	Respondents were interviewed face-to-face in Lithuanian; the item and option text is the authors’ own English rendering from the data dictionary (statements printed in capitals; option 4 ‘I agree more quickly’ means ‘tend to agree’), as the Lithuanian wording was not published. The table combines two blocks: views of media science reporting (m40-m42) and trust in scientists (m75-m79). m78 and m79 are negatively worded and stored unreversed, and answer code 9 (N/A) is treated as missing.
sleboda_2021_risk_benefit	For the five New items the anchor direction differs by technology, following the study’s SPSS value labels and the data: for gene technology (plant breeding, stem cells) and pesticides 1 = completely new and 11 = very familiar, while for E-number food additives and vaccination 1 = very familiar and 11 = completely new; the study’s English supplement states 1 = old, 11 = new for all five. Item and technology wording is the study’s English rendering of a Swedish administration
smirnov_2025_enrollment_motives	These eleven items are the options of one select-any question (‘What were your goals for enrolling into the doctorate?’), coded 1 = ticked, 0 = not ticked. The wording is the authors’ English rendering, from the paper’s Figure 1, of a survey administered in Russian; the Russian wording is not published. The deferment option’s ‘(male only)’ comes from the figure, although 3 women in the data selected it.
smpi_lorenzoluaces_2020_gad7	The GAD-7 item and response-option wording here is the published instrument’s; the study’s paper describes its 0-3 scale endpoints as “none of the days” and “all the days” rather than the GAD-7’s “Not at all” and “Nearly every day”, and no source shows the labels respondents actually saw
smpi_lorenzoluaces_2020_phq9	The PHQ-9 item and response-option wording here is the published instrument’s; the study’s paper describes its 0-3 scale endpoints as “none of the days” and “all the days” rather than the PHQ-9’s “Not at all” and “Nearly every day”, and no source shows the labels respondents actually saw
socialstereotype_hughes_2025_judgement	IRW has 6 items for affiliation but the codebook only has 5 items
song_2023_rses	Respondents were Chinese medical students, but neither the paper nor its data deposit publishes the questionnaire wording (or states its language outright), so IRW ships the Rosenberg Self-Esteem Scale’s English original, with language recorded as Chinese. Responses are stored already reverse-scored, so higher is always higher self-esteem: for RSES3, RSES5, RSES9 and RSES10 a 4 means Strongly disagree. RSES8 (‘I wish I could have more respect for myself’) is an exception to the paper’s own keying: the paper lists it as negatively rated, but its stored values behave as unreversed agreement (91% at 3-4, clustering with the positive items), so it ships with 4 = Strongly agree; the study published no anchors, so this direction is inferred from the data
song_2025_cs	Respondents were Chinese hotel frontline employees and the questionnaire was back-translated into Chinese, but neither the paper nor its data deposit publishes the Chinese wording, so IRW ships the English the paper itself prints in its Table 3, with language recorded as Chinese. The study reports only that a five-point Likert scale was used and never prints the anchor labels, so response options carry no text
song_2025_er	Item text for song_2025_er is the English the paper printed, not the Chinese respondents answered (the study was administered in China with back translation and no Chinese wording is published); and in the paper’s Table 3 two of the five items appear as bare descriptors rather than full items – ER2 reads ‘Level of satisfaction with job’ and ER3 ‘Experience: Advancement Opportunities’, transcribed as printed.
song_2025_pj	The source article’s Table 3 prints the same sentence, “Are your department’s procedures consistent with ethical standards?”, against both PJ3 and PJ4, so only three of this scale’s four items have published wording; IRW attributes that sentence to PJ3 and ships PJ4’s item_text blank rather than repeating it. No response anchors are shipped: the study reports only that a five-point Likert scale was used and never prints the labels.
song_2025_pwb	The item wording shipped for this table is the English printed in the source article; the questionnaire was administered in Chinese (back-translated) and the Chinese original is published nowhere in the paper or its supplement. No response-option labels are shipped because the study reports only that a five-point Likert scale was used.
song_2025_sc	Item text for song_2025_sc is the English the source article printed, not the Chinese respondents answered (the study was administered in China with back translation and no Chinese wording is published anywhere in the paper or its supplement), and no response-option labels are shipped because the study reports only that a five-point Likert scale was used. Note a direction mismatch in the response data: SC1-SC5 are stored reverse-keyed relative to the positive wording shipped for them – they correlate -0.34 to -0.53 with SC6 (‘There is a true commitment to service…’), and -0.43 to -0.53 with the same respondents’ positively worded procedural-justice items – so for SC1-SC5 a higher resp means less agreement with the sentence in item_text, while SC6 runs the natural way
song_2025_wwb	The item wording shipped for this table is the English printed in the source article; the questionnaire was administered in Chinese (back-translated) and the Chinese original is published nowhere in the paper or its supplement. No response-option labels are shipped because the study reports only that a five-point Likert scale was used.
spain_2012_entrepreneurship_barriers	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 2938 in Spanish only; the administered Spanish is in the base fields. Note also that resp runs from 1 ‘Muy de acuerdo’ to 4 ‘Muy en desacuerdo’, so a higher value means more disagreement, while the question’s lead-in names the ends ‘totalmente de acuerdo’ and ‘totalmente en desacuerdo’ (both as CIS printed them); ‘N.S.’ and ‘N.C.’ (about 10% on p1702 and p1703) are dropped to missing by the IRW processing script.
spain_2012_entrepreneurship_economy	The English item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 2938 in Spanish only; the administered Spanish is in the base fields. Note also that resp runs 1 ‘Mejor’ (better), 2 ‘Igual’ (the same), 3 ‘Peor’ (worse), so a higher value is more pessimistic, and that ‘N.S.’ and ‘N.C.’ (8% on p5) are dropped to missing by the IRW processing script.
spain_2012_entrepreneurship_education	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 2938 in Spanish only; the administered Spanish is in the base fields, with CIS’s printed tense alternatives (‘Te ayuda/ayudó’, ‘Te da/dio’) kept so one wording serves students and former students. Note also that resp runs from 1 ‘Muy de acuerdo’ to 4 ‘Muy en desacuerdo’, so a higher value means more disagreement, and that ‘N.S.’ and ‘N.C.’ (about 3% per item) are dropped to missing by the IRW processing script.
spain_2012_entrepreneurship_friends	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 2938 in Spanish only; the administered Spanish is in the base fields. Note also that resp runs from 1 ‘Más de la mitad’ to 3 ‘Menos de la mitad’, so a higher value means fewer friends from the respondent’s own group; that ‘N.S.’ and ‘N.C.’ (about 20% on political ideology, 13% on religious beliefs) are dropped to missing by the IRW processing script; and that the question’s lead-in names ‘sexo’ among the characteristics, although CIS printed and recorded no row for it.
spain_2012_entrepreneurship_image	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 2938 in Spanish only; the administered Spanish is in the base fields. Note also that resp takes only the values 1 ‘Positiva’ and 3 ‘Negativa’: respondents could volunteer a middle ‘Ni positiva ni negativa’ answer (CIS code 2, not read aloud), which the IRW processing script drops to missing along with ‘N.S.’ and ‘N.C.’. That neutral answer is 16% of all responses (6-22% per occupational group), so proportions computed from this table exclude a sizeable group.
spain_2012_entrepreneurship_motives	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 2938 in Spanish only; the administered Spanish is in the base fields, with CIS’s printed tense alternatives (‘Tenías/tienes’, ‘Era/es’) kept so one wording serves current and past business owners. Note also that the question was asked only of the 140 respondents (of 1,437) who have, had or had started setting up a business of their own, so the table holds 139 people, and that resp runs from 1 ‘Muy importante’ to 4 ‘Nada importante’, so a higher value means less important.
spain_2012_entrepreneurship_unemployment	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 2938 in Spanish only; the administered Spanish is in the base fields. Note also that resp runs from 1 ‘Muy importante’ to 4 ‘Nada importante’, so a higher value means the cause was rated less important, and that ‘N.S.’ and ‘N.C.’ (1-6% per item) are dropped to missing by the IRW processing script.
spain_2012_entrepreneurship_values	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 2938 in Spanish only; the administered Spanish is in the base fields. Note also that the 0-10 scale was labelled on the show card at its two ends only (0 ‘Muy poco importante’, 10 ‘Muy importante’), so values 1 to 9 carry no option text, and that higher values mean more important, the opposite direction to the 1-4 importance questions in the sibling spain_2012_entrepreneurship tables.
spain_2012_gender_acceptability	The English item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 2968 in Spanish only; the administered Spanish is in the base fields. Item p7 is a separate question (P.7, on violence by a man towards his wife or partner) whose first option reads ‘Algo inevitable que siempre ha existido’ rather than P.5’s ‘Algo inevitable’.
spain_2012_gender_beliefs	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 2968 in Spanish only; the administered Spanish is in the base fields. Codes run 1 ‘Muy de acuerdo’ to 4 ‘Nada de acuerdo’, so a higher value means less agreement. Item p901 follows CIS’s variable label (‘entre los extranjeros’), correcting a typo in the printed questionnaire (‘entre lo extranjeros’).
spain_2012_gender_campaigns	The English item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 2968 in Spanish only; the administered Spanish is in the base fields. Item p21 is a follow-up that refers back to the campaigns asked about in p20 (‘estas campañas’); codes are 1 ‘Sí’, 2 ‘No’.
spain_2012_gender_causes	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 2968 in Spanish only; the administered Spanish is in the base fields. Codes are 1 ‘Sí’ (considered a cause), 2 ‘No’. Item wording follows the printed questionnaire, which differs slightly from CIS’s variable labels for p1202 (‘El abuso del alcohol’ against ‘El abuso de alcohol’) and p1211 (a comma), and p1207’s printed typo ‘ee las relaciones’ is corrected to ‘de las relaciones’.
spain_2012_gender_complaints	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 2968 in Spanish only; the administered Spanish is in the base fields. Codes run 1 ‘Muy de acuerdo’ to 4 ‘Nada de acuerdo’, so a higher value means less agreement.
spain_2012_gender_coordination	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 2968 in Spanish only; the administered Spanish is in the base fields. Codes run 1 ‘Muy de acuerdo’ to 4 ‘Nada de acuerdo’, so a higher value means less agreement; ‘N.S.’/‘N.C.’ answers are dropped to missing by the IRW processing script, which removes 33% of respondents on p1602 and 28% on p1604.
spain_2012_gender_couple	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 2968 in Spanish only; the administered Spanish is in the base fields. Codes run 1 ‘Muy importante’ to 4 ‘Nada importante’, so a higher value means less important.
spain_2012_gender_custody	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 2968 in Spanish only; the administered Spanish is in the base fields. Items p2701 and p2702 carry each other’s CIS variable names: the printed questionnaire’s column markers and the response patterns put the statements one way, while CIS’s codebook, and the published marginals computed from it, attach them the other way, and the IRW response data inherit the codebook’s names. The item text here follows the data: p2701 is the statement that a man who abuses his partner need not be a bad father (29% agree; correlates +0.48 to +0.52 with the two pro-father items) and p2702 is that custody should be removed after a final conviction (92% agree; -0.28 to -0.31). Codes run 1 ‘Muy de acuerdo’ to 4 ‘Nada de acuerdo’.
spain_2012_gender_prevalence	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 2968 in Spanish only; the administered Spanish is in the base fields. Codes run 1 ‘Muy extendidos’ to 4 ‘Nada extendidos’, so a higher value means abuse is judged less widespread.
spain_2012_gender_punishment	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 2968 in Spanish only; the administered Spanish is in the base fields.
spain_2012_gender_situation	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 2968 in Spanish only; the administered Spanish is in the base fields. Codes are 1 ‘Mejor’, 2 ‘Igual’, 3 ‘Peor’, describing women’s situation relative to men’s, so a higher value means women are judged worse off.
spain_2012_gender_vulnerability	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 2968 in Spanish only; the administered Spanish is in the base fields. Codes are 1 ‘Sí’, 2 ‘No’, so 1 means the group is judged more vulnerable.
spain_2012_reality_conversation	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 2973 in Spanish only; the administered Spanish is in the base fields. Note also: codes run 1 = Con mucha frecuencia (very often) to 4 = Practicamente nunca (practically never), so higher means less political talk; ‘No recuerda’ and no-answer are stored as missing.
spain_2012_reality_media	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 2973 in Spanish only; the administered Spanish is in the base fields. Note also: codes run 1 = Todos los dias (every day) to 5 = Nunca (never), so higher means less frequent media use; don’t-know and no-answer are stored as missing.
spain_2012_reality_outlets	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 2973 in Spanish only; the administered Spanish is in the base fields. Note also: codes run 1 = Muy de izquierdas (very left-wing) to 5 = Muy de derechas (very right-wing); don’t-know and no-answer (very common here, 30-62% per outlet) are stored as missing; outlets were read in a rotated order per interview.
spain_2012_reality_participation	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 2973 in Spanish only; the administered Spanish is in the base fields. Note also: codes run 1 = done in the last twelve months, 2 = done in the more distant past, 3 = never done; no-answer is stored as missing.
spain_2012_reality_personality	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 2973 in Spanish only; the administered Spanish is in the base fields. Note also: codes run 1 = yes, completely to 5 = not at all (lower means the statement describes the respondent better); the unread middle option 3 (neither yes nor no), don’t-know and no-answer are stored as missing, so no respondent has a 3.
spain_2012_reality_press	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 2973 in Spanish only; the administered Spanish is in the base fields. Note also: codes are days per week (0 = none, 7 = every day); only respondents who said they read a newspaper at all were asked, so non-readers have no rows, and don’t-know/no-answer are stored as missing.
spain_2012_reality_radio	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 2973 in Spanish only; the administered Spanish is in the base fields. Note also: codes are days per week (0 = none, 7 = every day) of listening to each station’s news bulletins; only respondents who said they listen to radio news at all were asked, so non-listeners have no rows, and don’t-know/no-answer are stored as missing.
spain_2012_reality_situation	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 2973 in Spanish only; the administered Spanish is in the base fields. Note also: don’t-know/no-answer responses are stored as missing.
spain_2012_reality_television	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 2973 in Spanish only; the administered Spanish is in the base fields. Note also: codes are days per week (0 = none, 7 = every day) of watching each channel’s television news; only respondents who said they watch TV news at all were asked, so non-viewers have no rows, and don’t-know/no-answer are stored as missing.
spain_2013_defense_attractions	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 2998 in Spanish only; the administered Spanish is in the base fields, and p3607 uses the questionnaire’s ‘La camaradería y espíritu de equipo’ where the CIS data file labels it ‘La camaradería y el espíritu de equipo’. The stem’s ‘esa misma escala de 1 a 5’ refers back to the preceding question (in spain_2013_defense_conditions), and only the ends of that scale are labelled, 1 ‘Muy importante’ and 5 ‘Nada importante’, so values 2 to 4 carry no option text because the study printed none. The question was asked only of respondents aged 16 to 28 (464 of 2,479), so a missing cell means not asked, and ‘N.S.’/‘N.C.’ answers are dropped to missing by the IRW processing script.
spain_2013_defense_capability	The English item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 2998 in Spanish only; the administered Spanish is in the base fields. The two questions word their four-point scales differently (P18 ‘preparadas’, P20 ‘capacitados’), and P20’s question reads ‘muy, bastante, poco o muy poco capacitados’ while its printed fourth option is ‘Nada capacitados’; both are shipped as printed. ‘N.S.’/‘N.C.’ answers are dropped to missing by the IRW processing script, which removes about 16% of respondents on each item (413 and 386 of 2,479).
spain_2013_defense_children	The English item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 2998 in Spanish only; the administered Spanish is in the base fields. The Spanish options are identical on all four questions, but the English is gendered to match each question (her for the daughter items p43 and p44, him for the son items p45 and p46). The questions were asked only of respondents aged 29 or older (2,015 of 2,479; those aged 16 to 28 answered a different block), and they are hypothetical for respondents without a daughter or son rather than filtered on having one. A volunteered, not-read-aloud ‘(No leer) Ni uno ni lo otro’ answer (CIS code 3, 392 to 412 respondents per item) is dropped to missing by the IRW processing script along with ‘N.S.’/‘N.C.’, so only ‘Le animaría’ and ‘Se lo desaconsejaría’ remain.
spain_2013_defense_conditions	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 2998 in Spanish only; the administered Spanish is in the base fields, and p3507 and p3508 use the questionnaire’s ‘Administración’ where the CIS data file abbreviates it ‘Admón.’. Only the ends of the 1-5 scale are labelled, 1 ‘Muy importante’ and 5 ‘Nada importante’, so values 2 to 4 carry no option text because the study printed none. The question was asked only of respondents aged 16 to 28 (464 of 2,479), so a missing cell means not asked, and ‘N.S.’/‘N.C.’ answers are dropped to missing by the IRW processing script.
spain_2013_defense_emotion	The English item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 2998 in Spanish only; the administered Spanish is in the base fields. p5, on the national anthem, is worded elliptically as a follow-on to p4’s question (‘¿Y cuando escucha el himno nacional….’) and is shipped as printed, not expanded. On p4 and p5 respondents could volunteer ‘(No leer) Depende del acto o la ceremonia de que se trate’ (CIS code 5, not read aloud), which the IRW processing script drops to missing along with ‘N.S.’/‘N.C.’.
spain_2013_defense_missions	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 2998 in Spanish only; the administered Spanish is in the base fields. The question asks about the armed forces’ ‘misiones en el exterior’ (missions abroad); the preceding questions, not in this table, spoke of international peace missions. p3006’s last word is shipped as ‘bienestar’, the spelling in the CIS data file, where the questionnaire misprints it ‘bienestrar’. ‘N.S.’/‘N.C.’ answers are dropped to missing by the IRW processing script, which removes 10% to 20% of the 2,479 respondents per item (487 on p3005, economic benefits).
spain_2013_defense_prestige	The English item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 2998 in Spanish only; the administered Spanish is in the base fields. p13 asks about the international prestige of a country in general and p14 about Spain. ‘N.S.’/‘N.C.’ answers (138 and 196 of 2,479 respondents) are dropped to missing by the IRW processing script.
spain_2013_defense_professions	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 2998 in Spanish only; the administered Spanish is in the base fields. The English does not reproduce the Spanish gendered ‘/a’ forms, and p102’s ‘Profesor/a//Maestro/a’ (shipped as the questionnaire prints it, doubled slash included) is rendered ‘Teacher//Schoolteacher’ because profesor and maestro name secondary and primary teachers. The 0-10 scale was labelled only at its ends, 0 ‘Muy mal’ and 10 ‘Muy bien’, so values 1 to 9 carry no option text because the study printed none; ‘N.S.’/‘N.C.’ answers are dropped to missing by the IRW processing script.
spain_2013_defense_resources	The English item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 2998 in Spanish only; the administered Spanish is in the base fields. ‘N.S.’/‘N.C.’ answers are dropped to missing by the IRW processing script, which removes about 30% of respondents on every item (743, 774 and 738 of 2,479).
spain_2013_defense_sacrifice	The English instruction, item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 2998 in Spanish only; the administered Spanish is in the base fields. Each item completes the stem ‘Quisiera que Ud. me dijera si se sacrificaría o arriesgaría su vida por….’, which is why p7a01 (‘Su patria, su nación, su país’) has no leading ‘Por’ while the other six do. The block was asked only of respondents who had first answered yes to a question not in this table, whether anything apart from their family was worth risking their life for (1,169 of 2,479); everyone else is coded ‘No procede’, which the IRW processing script drops along with ‘N.S.’/‘N.C.’, so a missing cell usually means not asked.
spain_2013_defense_threats	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 2998 in Spanish only; the administered Spanish is in the base fields, and p907 keeps the questionnaire’s ‘químicas, Biológicas,…)’ where the CIS data file reads ‘químicas, biológicas, etc.)’. The question asks how important each risk or threat is for Spain’s security. The 0-10 scale was labelled only at its ends, 0 ‘Nada importante’ and 10 ‘Muy importante’, so values 1 to 9 carry no option text because the study printed none. ‘N.S.’/‘N.C.’ answers are dropped to missing by the IRW processing script, which removes 5% to 20% of the 2,479 respondents per item (508 on p903, cyberattacks, and 492 on p909, espionage).
spain_2013_services_complaints	The English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only; p27c’s code 3 (‘still being processed’) and CIS’s routing code 0 (‘No procede’, not asked) are stored as missing in the response table, so p27c is 1 Yes / 2 No and p27d runs 1-5; on p27d higher values mean less satisfied, with 3 (‘Indiferente’) an unprompted category not read to respondents.
spain_2013_services_functioning	The English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only; the three items use different response scales (p1 four points, p2 five points, p301-p309 three points), and the middle options ‘Igual’ (p2) and ‘Siguen igual’ (p301-p309) were marked ‘(NO LEER)’, i.e. accepted and coded but not read aloud to respondents.
spain_2013_services_importance	The English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only; on p6a, resp 0 (‘No procede’) is a CIS routing code for the 1525 of 2484 respondents who were not asked the question (only those who chose ‘it is essential that the quality of public services be evaluated’ at the preceding P.6 were asked), which CIS declares missing but the IRW table retains as a value, so it should not be analysed as a response.
spain_2013_services_inpersonsat	The English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only; all items refer to the respondent’s most recent in-person visit to a public office in the previous 12 months, and respondents who reported no such visit were not asked them but appear in the response table with resp 0 (‘No procede’, a CIS not-applicable code that should be treated as missing, 591 respondents on every item); response scales differ by item (p12 and p13 five points with an unread middle option, p1401-p1409 four points with no middle, so resp 3 means ‘Poco satisfecho/a’ there but ‘Indiferente’ on p12; p15 three points), and the middle options marked ‘(NO LEER)’ were accepted and coded but not read aloud.
spain_2013_services_internet	The English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only; p19 was not asked of the 863 of 2484 respondents who had not used the Internet in the past year, and they are missing; p2101-p2110 are a mark-all-that-apply list asked only of the 1134 who answered ‘Sí’ to p19, coded 1 = mentioned, 0 = asked but not mentioned (CIS prints no wording for this, so its option text is blank), and missing = not asked, or no answer to the whole question.
spain_2013_services_internetout	The English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only; all items refer to the respondent’s most recent visit to a public administration website in the past 12 months and were asked only of the 1134 of 2484 respondents who answered ‘Sí’ at P.19, so resp 0 (‘No procede’, a CIS not-applicable code that should be treated as missing) marks the 1350 respondents not asked on every item, plus on p22b the 87 who got nothing they were looking for (p22 = 4) and were routed past it; p23 is a five-point scale with an unread middle option (‘Indiferente’, marked ‘(NO LEER)’), so resp 3 = ‘Indiferente’ there but ‘Poco satisfecho/a’ on the four-point p2401-p2406.
spain_2013_services_offices	The English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only; only the scale end points (1 ‘Nada importante’, 10 ‘Muy importante’) are labelled in the source, so option text for 2-9 is blank.
spain_2013_services_othercontact	In p16a, p17a, p18a and p18b resp 0 is the CIS routing code ‘No procede’ (not asked because of the answer to the preceding filter question: 2219, 1576, 1622 and 1622 of 2484 respondents), which the processing script kept as a response instead of setting it to missing; exclude resp 0 before analysis.
spain_2013_services_purpose	The English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only; this mark-all-that-apply question was asked only of the 1893 of 2484 respondents who had visited a public office in person in the past 12 months, and is coded 1 = mentioned, 0 = asked but not mentioned (CIS prints no wording for this, so its option text is blank), and missing = not asked, or no answer to the whole question.
spain_2013_services_services	The English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only.
spain_2014_debate_agreement	The English instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3016 in Spanish only; the administered Spanish is in the base fields, and the items are the names of 17 political leaders, left unchanged in English (item 9 uses the questionnaire’s ‘Xabier Mikel Errekondo’, which the CIS data file labels ‘Mikel Errekondo’). The stem opens ‘Y’, continuing a preceding question, not in this table, that rated each leader’s speech in the 2014 State of the Nation debate. Each leader was asked about only by respondents who followed the debate and recalled that leader’s intervention, so the number answering runs from 176 (p13a_14) to 1,035 (p13a_1) and a missing cell means not asked, not item nonresponse. ‘N.S.’/‘N.C.’ answers are dropped to missing by the IRW processing script; ‘N.S.’ alone accounts for 242 to 313 answers per leader, except Rajoy (37) and Rubalcaba (60).
spain_2014_debate_debate	The English item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3016 in Spanish only; the administered Spanish is in the base fields (‘conflictivo’ is rendered ‘confrontational’). P12’s wording varied by fieldwork day: the interviewer read ‘está siendo’ (is being) on Wednesday and ‘ha sido’ (has been) on Thursday, and the item text keeps both, in brackets, as the questionnaire prints them (‘[MIÉRCOLES está siendo, JUEVES ha sido]’). P11 opens ‘Y,’, continuing a preceding question not in this table. The questions were asked only of respondents who knew of the February 2014 State of the Nation debate and had followed it or received information about it (1,087 of 1,602), and ‘N.S.’/‘N.C.’ answers are dropped to missing by the IRW processing script.
spain_2014_debate_general	The English item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3016 in Spanish only; the administered Spanish is in the base fields. Both questions refer to ‘estos debates’, the State of the Nation debates introduced by the survey’s preceding question, which is not in this table. P2’s option 3 is shipped as ‘Poco interesantes’, the plural used in the CIS data file and in the question as read, where the questionnaire’s option list prints ‘Poco interesante’. ‘N.S.’/‘N.C.’ answers are dropped to missing by the IRW processing script.
spain_2014_debate_government	The English instruction, item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3016 in Spanish only; the administered Spanish is in the base fields. The questions ask about the message conveyed by the President of the Government, Mariano Rajoy, in the 2014 State of the Nation debate, not about the Government in general, and were asked only of respondents who followed the debate and recalled Rajoy’s intervention (1,075 of 1,602), so a missing cell means not asked. The scale runs 1 ‘Mucho’ to 4 ‘Nada’, and ‘N.S.’/‘N.C.’ answers are dropped to missing by the IRW processing script.
spain_2014_debate_leaders	The English instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3016 in Spanish only; the administered Spanish is in the base fields, and the items are the names of 17 political leaders, left unchanged in English (item 9 uses the questionnaire’s ‘Xabier Mikel Errekondo’, which the CIS data file labels ‘Mikel Errekondo’). Note also that the response values run 1, 2, 4, 5 with no option row for 3: the interviewer did not read out a middle ‘(No leer) Regular’ answer (CIS code 3) or ‘(No leer) No conoce al líder, no lo recuerda’ (code 7), and the IRW processing script drops both to missing along with 8 ‘N.S.’ and 9 ‘N.C.’. The questions were asked only of respondents who followed the 2014 State of the Nation debate, and because not knowing the leader is dropped, the number answering runs from 141 (p13_14) to 907 (p13_1).
spain_2014_debate_media	The English instruction, item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3016 in Spanish only; the administered Spanish is in the base fields. The questionnaire’s stem and its grid rows word each activity slightly differently (for example ‘ve un programa de noticias políticas’ in the stem, ‘Ve noticias políticas en la televisión’ in the row), and both are shipped as printed. The few ‘N.C.’ answers are dropped to missing by the IRW processing script.
spain_2014_debate_opposition	The English instruction, item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3016 in Spanish only; the administered Spanish is in the base fields. The items ask whether Alfredo Pérez Rubalcaba, in the 2014 State of the Nation debate, conveyed that the PSOE, his party, has each quality; the subject of the four statements is the PSOE, not Rubalcaba personally. The stem opens ‘Y’, continuing a preceding question about Mariano Rajoy that is not in this table. The questions were asked only of respondents who followed the debate and recalled Rubalcaba’s intervention (1,074 of 1,602), so a missing cell means not asked; the scale runs 1 ‘Mucho’ to 4 ‘Nada’, and ‘N.S.’/‘N.C.’ answers are dropped to missing by the IRW processing script.
spain_2014_debate_rajoy	The English instruction, item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3016 in Spanish only; the administered Spanish is in the base fields, and each item is a short phrase completing the stem (whether Rajoy ‘demostró’ each quality). The questions were asked only of respondents who followed the 2014 State of the Nation debate and recalled Mariano Rajoy’s intervention (1,075 of 1,602), so a missing cell means not asked, and ‘N.S.’/‘N.C.’ answers are dropped to missing by the IRW processing script.
spain_2014_debate_rubalcaba	The English instruction, item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3016 in Spanish only; the administered Spanish is in the base fields. The stem is elliptical: it opens ‘Y, …’ and continues the parallel question about Mariano Rajoy (in spain_2014_debate_rajoy), whose full wording is not repeated here, and each item is a short phrase completing it. The questions were asked only of respondents who followed the 2014 State of the Nation debate and recalled Alfredo Pérez Rubalcaba’s intervention (1,074 of 1,602), so a missing cell means not asked, and ‘N.S.’/‘N.C.’ answers are dropped to missing by the IRW processing script.
spain_2014_family_activities	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3032 in Spanish only; the administered Spanish is in the base fields. Note also that P.21g was asked only of respondents living with at least one child aged 25 or under, so missing cells mostly mean the question was not asked.
spain_2014_family_adoption	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3032 in Spanish only; the administered Spanish is in the base fields. Note also that the response values run 1, 2, 4, 5 with no option row for 3: respondents could volunteer a middle ‘ni de acuerdo ni en desacuerdo’ answer (CIS code 3, not read aloud), which the IRW processing script drops to missing along with 8 ‘N.S.’ and 9 ‘N.C.’.
spain_2014_family_care	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3032 in Spanish only; the administered Spanish is in the base fields. Note also that the response values run 1, 2, 4, 5 with no option row for 3: respondents could volunteer a middle ‘ni de acuerdo ni en desacuerdo’ answer (CIS code 3, not read aloud), which the IRW processing script drops to missing along with 9 ‘N.C.’. Items p504 and p505 follow the questionnaire’s wording, which differs from the CIS variable labels only in gender-inclusive forms.
spain_2014_family_change	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3032 in Spanish only; the administered Spanish is in the base fields. The question asks whether family use of new technologies (Internet, mobile phone) increases or decreases each aspect, which the item rows alone do not say. Respondents could volunteer ‘stays the same’ (CIS code 3, not read aloud), which the IRW processing script drops to missing along with 8 ‘N.S.’ and 9 ‘N.C.’, so resp takes only 1 (increases) and 2 (decreases).
spain_2014_family_chores	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3032 in Spanish only; the administered Spanish is in the base fields. Note also that P.8d was asked only of respondents living with their partner (married or not), and that the answer ‘a third person does it’ (CIS code 6, on the show card) is dropped to missing by the IRW processing script, so resp runs only from 1 (always the woman) to 5 (always the man).
spain_2014_family_conflicts	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3032 in Spanish only; the administered Spanish is in the base fields. Note also that P.21d was asked only of respondents living with at least one child aged 12 to 25, and that resp runs from 1 (a serious problem) to 3 (not a problem).
spain_2014_family_families	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3032 in Spanish only; the administered Spanish is in the base fields. Note also that the response values run 1, 2, 4, 5 with no option row for 3: respondents could volunteer a middle ‘ni de acuerdo ni en desacuerdo’ answer (CIS code 3, not read aloud), which the IRW processing script drops to missing along with 8 ‘N.S.’ and 9 ‘N.C.’. Items p1208 and p1210 follow the questionnaire’s wording, which differs from the CIS variable labels only in gender inflection.
spain_2014_family_housework	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3032 in Spanish only; the administered Spanish is in the base fields. Note also that P.40a was asked only of respondents in paid work, and each item asks how often the respondent has difficulty combining their job with the task, not how often they do it: 3 ‘Nunca’ means never has difficulty. The answer ‘does not do these tasks’ (CIS code 4) is dropped to missing by the IRW processing script.
spain_2014_family_importance	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3032 in Spanish only; the administered Spanish is in the base fields. Note also that resp runs from 1 ‘Muy importante’ to 4 ‘Nada importante’, so lower values mean more important.
spain_2014_family_marriage	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3032 in Spanish only; the administered Spanish is in the base fields. Note also that resp runs from 1 ‘Mucho’ to 4 ‘Nada’, so lower values mean a reason is seen as more influential in people marrying rather than cohabiting.
spain_2014_family_matrimony	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3032 in Spanish only; the administered Spanish is in the base fields. Note also that the response values run 1, 2, 4, 5 with no option row for 3: respondents could volunteer a middle ‘ni de acuerdo ni en desacuerdo’ answer (CIS code 3, not read aloud), which the IRW processing script drops to missing along with 8 ‘N.S.’ and 9 ‘N.C.’. The items complete the stem ‘… casarse/contraer matrimonio es …’; p1003 and p1005 follow the questionnaire’s wording, which differs from the CIS variable labels only in gender inflection.
spain_2014_family_obstacles	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3032 in Spanish only; the administered Spanish is in the base fields. Note also that the response values run 1, 2, 4, 5 with no option row for 3: respondents could volunteer a middle ‘ni de acuerdo ni en desacuerdo’ answer (CIS code 3, not read aloud), which the IRW processing script drops to missing along with 8 ‘N.S.’ and 9 ‘N.C.’.
spain_2014_family_reproduction	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3032 in Spanish only; the administered Spanish is in the base fields. Note also that respondents could volunteer ‘Depende de los casos’ (CIS code 3, not read aloud), which the IRW processing script drops to missing along with 8 ‘N.S.’ and 9 ‘N.C.’, so resp is binary: 1 ‘Mas bien a favor’, 2 ‘Mas bien en contra’.
spain_2014_family_satisfaction	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3032 in Spanish only; the administered Spanish is in the base fields. Note also that resp runs from 1 ‘Muy satisfecho/a’ to 4 ‘Nada satisfecho/a’, and that the four items that could be inapplicable (work p201, household income p202, salary p207, partner p210) have ‘no procede’ answers dropped to missing by the IRW processing script, so their n is smaller by design.
spain_2014_volunteering_activity	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3039 in Spanish only; the administered Spanish is in the base fields. Note also: despite the table name and the IRW dictionary description (‘active participation’), these items are CIS question P.10a, which asks whether each association the respondent belongs or belonged to is/was specifically a youth association (1 = Si, 2 = No); asked only of current or former members of that association type.
spain_2014_volunteering_membership	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3039 in Spanish only; the administered Spanish is in the base fields. Note also: option_text follows the questionnaire’s second-person wording, while the show card respondents were handed words the options in the first person (‘SI PERTENEZCO’ / ‘YA NO PERTENEZCO, PERO PERTENECI’ / ‘NUNCA HE PERTENECIDO’).
spain_2014_volunteering_motives	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3039 in Spanish only; the administered Spanish is in the base fields. Note also: asked of all respondents (not only volunteers) about the motivations someone may have for volunteering, not the respondent’s own motives; resp 1 = Mucha (a lot of importance) to 4 = Ninguna (none), so higher values mean less importance.
spain_2014_volunteering_profile	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3039 in Spanish only; the administered Spanish is in the base fields. Note also: asked of all respondents (not only volunteers) how they picture people who volunteer, one yes/no answer per characteristic (the characteristics are not mutually exclusive, e.g. Hombres and Mujeres are separate items); resp 1 = Si (yes), 2 = No.
spain_2014_volunteering_satisfaction	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3039 in Spanish only; the administered Spanish is in the base fields. Note also: resp 1 = Muy satisfecho/a (very satisfied) to 4 = Nada satisfecho/a (not at all satisfied), so higher values mean less satisfaction; p3 is overall life satisfaction (P.3) and p501-p511 are domain satisfaction (P.5); P.5 had no not-applicable option, so p503 (your work) was also answered by 470 respondents who were not working, 313 of them choosing Nada satisfecho/a.
spain_2014_volunteering_solidarity	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3039 in Spanish only; the administered Spanish is in the base fields. Note also: resp 1 = Mucho (a lot) to 4 = Nada (not at all), so higher values mean less agreement with the statement about solidarity.
spain_2015_immigration_acceptability	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3119 in Spanish only; the administered Spanish is in the base fields. Codes run 1 ‘Muy aceptable’ to 4 ‘Nada aceptable’, so a higher value means less acceptable; ‘N.S.’ and ‘N.C.’ answers (5-10% of respondents per item) are dropped to missing by the IRW processing script.
spain_2015_immigration_assistance	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3119 in Spanish only; the administered Spanish is in the base fields. The question asks how much protection from the State each group should receive; codes run 1 ‘Mucha’ to 4 ‘Ninguna’, and ‘N.S.’/‘N.C.’ answers, 5.9% (pensioners) to 12.6% (immigrants) of respondents per item, are dropped to missing by the IRW processing script.
spain_2015_immigration_contact	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3119 in Spanish only; the administered Spanish is in the base fields. Items p26a-p26c are the interviewer’s elliptical follow-ups to p26 (e.g. ‘¿Y entre todos sus amigos y amigas?’), so each asks how many people in that group are immigrants; codes run 1 ‘Muchos’ to 4 ‘Ninguno’. The not-applicable code ‘No procede’ is dropped to missing by the IRW processing script along with ‘N.S.’/‘N.C.’, which removes 42% of respondents from p26b (co-workers or fellow students), mostly people with no workplace or place of study.
spain_2015_immigration_criteria	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3119 in Spanish only; the administered Spanish is in the base fields. The 0-10 importance scale was labelled at its ends only (0 ‘Nada importante’, 10 ‘Muy importante’), so resp 1-9 carry no option text; ‘N.S.’/‘N.C.’ answers (3-5% per item) are dropped to missing by the IRW processing script.
spain_2015_immigration_distance	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3119 in Spanish only; the administered Spanish is in the base fields. Codes run 1 ‘Acepta/ría’ to 3 ‘Rechazaría’; respondents could also volunteer ‘(NO LEER) Depende’ (it depends), which the interviewer did not read aloud, and the IRW processing script drops it to missing along with ‘N.S.’/‘N.C.’, so 90-254 ‘Depende’ answers per item are absent from the table and have no option row.
spain_2015_immigration_diversity	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3119 in Spanish only; the administered Spanish is in the base fields. Each item is a short completion of the shared stem, which asks how positive or negative it is that Spanish society is made up of people of different countries, cultures, religions or skin colours; the 0-10 scale was labelled at its ends only (0 ‘Muy negativo’, 10 ‘Muy positivo’), so resp 1-9 carry no option text, and ‘N.S.’/‘N.C.’ answers are dropped to missing by the IRW processing script.
spain_2015_immigration_education	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3119 in Spanish only; the administered Spanish is in the base fields. Codes run 1 ‘Muy de acuerdo’ to 4 ‘Muy en desacuerdo’; ‘N.S.’/‘N.C.’ answers (6-18% of respondents per item) are dropped to missing by the IRW processing script. Item p1501 follows CIS’s variable label (‘en las escuelas’) where the printed questionnaire has the typo ‘en la escuelas’.
spain_2015_immigration_health	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3119 in Spanish only; the administered Spanish is in the base fields. All four statements are worded unfavourably to immigrants and codes run 1 ‘Muy de acuerdo’ to 4 ‘Muy en desacuerdo’, so a low value is the more negative attitude; ‘N.S.’/‘N.C.’ answers, 4.1-15.2% of respondents per item, are dropped to missing by the IRW processing script.
spain_2015_immigration_labour	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3119 in Spanish only; the administered Spanish is in the base fields. Codes run 1 ‘Muy de acuerdo’ to 4 ‘Muy en desacuerdo’ and are stored as recorded, but the items point in opposite directions: p2101 and p2102 are worded favourably to immigrants and p2103-p2105 unfavourably. ‘N.S.’/‘N.C.’ answers, 3.5-12.6% of respondents per item, are dropped to missing by the IRW processing script.
spain_2015_immigration_proximity	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3119 in Spanish only; the administered Spanish is in the base fields. Items p27a (friends) and p27b (co-workers or fellow students) follow the printed CIS questionnaire, whose order the data confirm; CIS’s own SPSS syntax and .sav file label these two variables the other way round, so do not relabel p27a/p27b from those files. Items p27a-p27c are the interviewer’s elliptical follow-ups to p27, each asking how many people in that group are Roma, with codes 1 ‘Muchos’ to 4 ‘Ninguno’. The not-applicable code ‘No procede’ is dropped to missing by the IRW processing script along with ‘N.S.’/‘N.C.’, which removes 40% of respondents from p27b.
spain_2015_immigration_punishment	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3119 in Spanish only; the administered Spanish is in the base fields. Codes run 1 ‘En todos los casos’ to 4 ‘En ningún caso’; respondents could also volunteer ‘(NO LEER) Depende’ (it depends), which the IRW processing script drops to missing along with ‘N.S.’/‘N.C.’, removing about 15% of respondents per item (about 10% for ‘Depende’ alone), so those who said it depends are absent rather than placed on the scale.
spain_2015_immigration_rights	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3119 in Spanish only; the administered Spanish is in the base fields. Codes are 1 ‘Sí’, 2 ‘No’; ‘N.S.’/‘N.C.’ answers, 4.1-12.1% of respondents per item (most for Spanish nationality), are dropped to missing by the IRW processing script.
spain_2015_immigration_roma	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3119 in Spanish only; the administered Spanish is in the base fields. Codes run 1 ‘Acepta/ría’ to 3 ‘Rechazaría’; the volunteered ‘(NO LEER) Depende’ (it depends) answer is dropped to missing by the IRW processing script along with ‘N.S.’/‘N.C.’, removing 6.4-12.3% of respondents per item. Item p2501 follows the questionnaire (‘…viven muchas personas gitanas’); CIS’s variable label omits ‘muchas’.
spain_2016_cooperation_awareness	The English in the _translated columns was produced by IRW (machine translation), not by CIS, which publishes this questionnaire in Spanish only. Each question also had a ‘N.C.’ (no answer) code, which the IRW processing script drops to missing.
spain_2016_cooperation_duty	The English in the _translated columns was produced by IRW (machine translation), not by CIS, which publishes this questionnaire in Spanish only. For p5, respondents could also volunteer ‘No estoy seguro/a’ (not sure; code 3, not read aloud, 8.4% of the sample), which the IRW processing script treats as missing, so only Sí (1) and No (2) appear.
spain_2016_cooperation_funding	The English in the _translated columns was produced by IRW (machine translation), not by CIS. ‘N.S.’ (don’t know) was the most common answer (60% for the Autonomous Community, 64% for the town council), and the IRW processing script drops it to missing, so the table holds only respondents who answered yes or no.
spain_2016_cooperation_interest	The English in the _translated columns was produced by IRW (machine translation), not by CIS. The table pools two related questions on parallel 4-point scales: p1 asks how much attention the respondent pays to news about other countries, while p201-p208 ask about interest in international news on each region (1 = Mucha/Mucho … 4 = Ninguna/Ninguno).
spain_2016_pain_beliefs	The English in the _translated columns was machine-translated by IRW from the Spanish CIS administered (CIS publishes this questionnaire in Spanish only); ‘N.S.’ (don’t know) and ‘N.C.’ (no answer) are dropped to missing by the IRW processing script, which removes up to 16% of respondents on some statements (395 of 2,464 on p25_4).
spain_2016_pain_coping	The English in the _translated columns was machine-translated by IRW from the Spanish CIS administered (CIS publishes this questionnaire in Spanish only); the IRW processing script drops code 7 ‘N.P. (No busca a otras personas)’ – respondents who say they do not turn to other people – to missing along with N.S./N.C., removing 120 of 2,314 eligible respondents on p21_1 (78 on p21_2, 56 on p21_3).
spain_2016_pain_difficulty	The English in the _translated columns was machine-translated by IRW from the Spanish CIS administered (CIS publishes this questionnaire in Spanish only); the IRW processing script dropped the volunteered middle answer ‘(NO LEER) Algo’ (code 3, 337 respondents on p16 and 317 on p17) to missing without renumbering, so resp runs 1, 2, 4, 5 and higher values mean less difficulty.
spain_2016_pain_effectiveness	The English in the _translated columns was machine-translated by IRW from the Spanish CIS administered (CIS publishes this questionnaire in Spanish only); the volunteered answer ‘some were effective and others not’ (17 respondents on p18a, 6 on p19a), ‘don’t know/don’t remember’ and ‘no answer’ are dropped to missing by the IRW processing script.
spain_2016_pain_fitness	The English in the _translated columns was machine-translated by IRW from the Spanish CIS administered (CIS publishes this questionnaire in Spanish only); ‘don’t know’ and ‘no answer’ (10 respondents on p29, 4 on p34) are dropped to missing by the IRW processing script.
spain_2016_pain_frequency	The English in the _translated columns was machine-translated by IRW from the Spanish CIS administered (CIS publishes this questionnaire in Spanish only); p8 asks how often the respondent feels pain of any kind, while p15 asks about the single most relevant pain they chose earlier (which may be a past pain), and the IRW processing script drops ‘don’t know/don’t remember’ (7 on p8, 22 on p15) and ‘no answer’ (23 on p15) to missing.
spain_2016_pain_health	The English in the _translated columns was machine-translated by IRW from the Spanish CIS administered (CIS publishes this questionnaire in Spanish only); on p4 (health compared with people of the same sex and age) the IRW processing script drops 106 ‘don’t know’ and 4 ‘no answer’ responses (4.5% of 2,464) to missing.
spain_2016_pain_support	The English in the _translated columns was machine-translated by IRW from the Spanish CIS administered (CIS publishes this questionnaire in Spanish only); the IRW processing script drops code 7 ‘N.P. (told nobody / has no family)’ to missing along with don’t know / no answer, removing 70, 63 and 67 respondents on p20_1, p20_2 and p20_3 who said they told nobody about the pain or have no family.
spain_2016_pain_wellbeing	The English in the _translated columns was machine-translated by IRW from the Spanish CIS administered (CIS publishes this questionnaire in Spanish only); the IRW processing script drops ‘not applicable’ answers to missing, which removes 927 of 2,464 respondents (38%) from p1_2 (satisfaction with work) and 417 (17%) from p1_6 (satisfaction with partner).
spain_2017_politics_agreement	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3184 in Spanish only; the administered Spanish is in the base fields (‘Hacienda’ is rendered ‘the Tax Agency’). The options are shipped in full, ‘Más bien de acuerdo’ and ‘Más bien en desacuerdo’, as the question reads them, where the questionnaire grid abbreviates them. ‘N.S.’/‘N.C.’ answers are dropped to missing by the IRW processing script (360 of 2,460 on p2302).
spain_2017_politics_attitudes	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3184 in Spanish only; the administered Spanish is in the base fields. The scale runs 1 ‘Muy de acuerdo’ to 5 ‘Muy en desacuerdo’, so a lower value means more agreement. Note also that the response values run 1, 2, 4, 5 with no option row for 3: respondents could volunteer a middle ‘(NO LEER) Ni de acuerdo, ni en desacuerdo’ answer (CIS code 3, not read aloud; 292 to 387 respondents per item) which the IRW processing script drops to missing along with 8 ‘N.S.’ and 9 ‘N.C.’, so 12% to 24% of respondents are missing on each item.
spain_2017_politics_burden	The English item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3184 in Spanish only; the administered Spanish is in the base fields (‘Regular’ is rendered ‘Moderate’). The four questions do not share a scale: p10 runs 1 ‘Mucho’ to 4 ‘Nada’, and p11 to p13 each have their own three options. On p13, which compares Spain’s tax burden with other European countries, 729 of 2,460 respondents (30%) volunteered ‘(NO LEER) No sabe, no puede comparar con otros países’ (CIS code 4); the IRW processing script drops it to missing, so far fewer respondents answer p13 than the other items.
spain_2017_politics_citizenship	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3184 in Spanish only; the administered Spanish is in the base fields. The 0-10 importance scale was labelled only at its ends, 0 ‘Nada importante’ and 10 ‘Muy importante’, so values 1 to 9 carry no option text because the study printed none, and ‘N.S.’/‘N.C.’ answers are dropped to missing by the IRW processing script.
spain_2017_politics_conscience	The English item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3184 in Spanish only; the administered Spanish is in the base fields. p16 asks how conscientious and responsible Spaniards are about paying taxes, and p17 asks the same of the respondent as an elliptical follow-up (‘¿Y Ud., personalmente, se considera a sí mismo/a…?’), shipped as printed, with its referent carried by the options. Both are coded 1 ‘Muy’ to 4 ‘Muy poco conscientes y responsables’, so a lower value means more conscientious. ‘N.S.’/‘N.C.’ answers are dropped to missing by the IRW processing script; p17 offered no ‘N.S.’ code.
spain_2017_politics_discussion	The English instruction, item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3184 in Spanish only; the administered Spanish is in the base fields, and each item is a short label (‘Amigos/as’, ‘Familiares’) completing the stem’s ‘…cuando se reúne con sus…?’. The scale runs 1 ‘A menudo’ to 4 ‘Nunca’, so a lower value means more discussion. The IRW processing script drops the not-read-aloud ‘(NO LEER) No procede’ code to missing along with ‘N.S.’/‘N.C.’, which removes 1,005 of 2,460 respondents (41%) from p2603, colleagues at work or study, against under 0.5% on the other two items.
spain_2017_politics_fraud	The English item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3184 in Spanish only; the administered Spanish is in the base fields (‘IVA’ is rendered ‘VAT’). The four questions have their own options, all coded so that 1 means the most fraud, the most income declared or the most effort against fraud. ‘N.S.’/‘N.C.’ answers are dropped to missing by the IRW processing script, which removes 364 of 2,460 respondents (15%) on p19 and 565 (23%) on p20, the questions about whether people declare all their income and VAT.
spain_2017_politics_services	The English instruction, item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3184 in Spanish only; the administered Spanish is in the base fields. The options are shipped as ‘Muy satisfactoriamente’ to ‘Nada satisfactoriamente’, the CIS data file’s labels, where the questionnaire grid prints only the bare ‘Muy’, ‘Bastante’, ‘Poco’, ‘Nada’ that complete the stem’s ‘funciona satisfactoriamente’; a lower value means better functioning. ‘N.S.’/‘N.C.’ answers are dropped to missing by the IRW processing script, from 27 of 2,460 respondents on p602 (health care) to 409 (17%) on p603 (pension management).
spain_2017_politics_spending	The English instruction, item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3184 in Spanish only; the administered Spanish is in the base fields, and each item is a bare service name completing the stem. The interviewer read the fifteen services in a rotated order that changed from interview to interview, so p801 to p815 is not the order any respondent heard, and the starting service is not in the table. The scale runs 1 ‘Demasiados’ to 3 ‘Muy pocos’, so a higher value means the service is seen as under-funded. ‘N.S.’/‘N.C.’ answers are dropped to missing by the IRW processing script, from 4.5% of respondents on p806 (Sanidad) to 24.8% on p813 (Cooperación al desarrollo).
spain_2017_politics_wellbeing	The English instruction, item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3184 in Spanish only; the administered Spanish is in the base fields. The table holds two questions with different anchors: p101 to p104 rate satisfaction with four life domains from 0 ‘Completamente insatisfecho/a’ to 10 ‘Completamente satisfecho/a’, and p2 rates personal happiness from 0 ‘Completamente infeliz’ to 10 ‘Completamente feliz’. Only those ends are labelled, so values 1 to 9 carry no option text because the study printed none, and ‘N.S.’/‘N.C.’ answers are dropped to missing by the IRW processing script.
spain_2018_housing_amenities	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3212 in Spanish only; the administered Spanish is in the base fields. The items are the questionnaire’s terse grid labels (for example ‘Garaje’, ‘Trastero’), each completing the shared stem ‘¿Dispone la vivienda de…?’. The few ‘N.C.’ answers (at most 3 per item) are dropped to missing by the IRW processing script.
spain_2018_housing_building	The English item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3212 in Spanish only; the administered Spanish is in the base fields, and it keeps the questionnaire’s printed alternatives for respondents in a detached house (‘/su vivienda’, ‘/de su vivienda’) and its unaccented ‘frio’ on P13_4. Each question also had 8 ‘N.S.’ and 9 ‘N.C.’ codes that the IRW processing script drops to missing, which removes 276 of 2,468 respondents on P13_7 (fire safety) and 91 on P13_6 (burglary security) against at most 18 on the other items.
spain_2018_housing_dwelling	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3212 in Spanish only; the administered Spanish is in the base fields. ‘N.S.’/‘N.C.’ answers (at most 21 of 2,468 respondents per item) are dropped to missing by the IRW processing script.
spain_2018_housing_measures	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3212 in Spanish only; the administered Spanish is in the base fields. The stem opens ‘Más concretamente’, referring back to a preceding question, not in this table, on whether the public administrations should adopt measures to encourage renting. Note also that the response values run 1, 2, 4, 5 with no option row for 3: respondents could volunteer a middle ‘(NO LEER) Ni a favor ni en contra’ answer (CIS code 3, not read aloud; 132 to 240 respondents per item) which the IRW processing script drops to missing along with 8 ‘N.S.’ and 9 ‘N.C.’.
spain_2018_housing_neighborhood	The English item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3212 in Spanish only; the administered Spanish is in the base fields. Each item also had 8 ‘N.S.’ and 9 ‘N.C.’ codes that the IRW processing script drops to missing (for example 68 on P3_3, cultural spaces, and 53 on P3_2, sports facilities).
spain_2018_housing_opinions	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3212 in Spanish only; the administered Spanish is in the base fields. Note also that the response values run 1, 2, 4, 5 with no option row for 3: respondents could volunteer a middle ‘(NO LEER) Ni de acuerdo ni en desacuerdo’ answer (CIS code 3, not read aloud) which the IRW processing script drops to missing along with 8 ‘N.S.’ and 9 ‘N.C.’. The volunteered midpoint is not rare here, 204 to 432 of 2,468 respondents per item, and ‘N.S.’ removes a further 384 on P25_2.
spain_2018_housing_ownership	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3212 in Spanish only; the administered Spanish is in the base fields. ‘N.S.’/‘N.C.’ answers (at most 10 respondents per item) are dropped to missing by the IRW processing script.
spain_2018_housing_problems	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3212 in Spanish only; the administered Spanish is in the base fields. The scale runs 1 ‘Muy importante’ to 4 ‘Nada importante’, so a higher value means a less important problem. ‘N.S.’/‘N.C.’ answers are dropped to missing by the IRW processing script, which removes a large share of respondents on some items (591 of 2,468 on P4_2, rental prices; 499 on P4_1, sale prices; 257 on P4_6, empty dwellings; 224 on P4_7, illegal occupation), so the set of respondents differs from item to item.
spain_2018_housing_rentals	The English item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3212 in Spanish only; the administered Spanish is in the base fields (‘Ajustados’ is rendered ‘Reasonable’). The three questions do not share a scale: P22 rates rent levels from 1 ‘Muy caros’ to 5 ‘Muy baratos’, P23 the trend from 1 ‘Están subiendo mucho’ to 5 ‘Están bajando mucho’, and P24 the prevalence of tourist rentals from 1 ‘Muchos’ to 4 ‘No hay’. ‘N.S.’/‘N.C.’ answers are dropped to missing by the IRW processing script, which removes 24% of the 2,468 respondents from P22, 27% from P23 and 13% from P24.
spain_2023_identity_authenticity	The English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only; the interviewer inserted each respondent’s own autonomous community where the text shows ‘(gentilicio de la C. A.)’; and CIS’s code 7 ‘Ninguna’ is stored as resp 4 by the IRW processing script, with ‘N.S.’ and ‘N.C.’ dropped to missing.
spain_2023_identity_centralism	The English instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3409 in Spanish only; the administered Spanish is in the base fields. Only the scale end points (1 and 10) carry verbal labels; values 2-9 were unlabelled in the questionnaire.
spain_2023_identity_europe	The English item and response-option wording for this table was produced by IRW, not by CIS, which publishes this questionnaire only in Spanish; the administered Spanish is in the base fields. Respondents could also volunteer ‘Ni le ha beneficiado ni le ha perjudicado’ (neither benefited nor harmed; CIS code 2, not read aloud, about 3% of the sample), which the IRW processing script treats as missing, so the response values are 1 and 3 with no option row for 2.
spain_2023_identity_identification	The English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only; respondents could also volunteer ‘Ni poco ni bastante’ (neither little nor quite a lot; CIS code 3, not read aloud, 1-3% per item), which the IRW processing script treats as missing along with ‘N.S.’ and ‘N.C.’, so the response values are 1, 2, 4 and 5 with no option row for 3.
spain_2023_identity_patriotism	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3409 in Spanish only; the administered Spanish is in the base fields. Responses are 1 (agree) and 3 (disagree) only: CIS code 2, an unprompted ‘neither agree nor disagree’, was set to missing when the table was built.
spain_2023_identity_pride	The English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only; in P.12 the interviewer inserted each respondent’s own autonomous community where the text shows ‘(gentilicio de la C. A.)’.
spain_2023_identity_state	The English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only; the administered Spanish is in the base fields.
spain_2023_identity_territorial	The English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only; for the autonomous-community item the interviewer also read out the name of the respondent’s own community after ‘Con la Comunidad Autónoma donde vive’, which is omitted here as it is in CIS’s own variable label; only the scale end points 0 and 10 carry verbal labels.
spain_2023_science_activities	The English in the _translated columns was machine-translated by IRW from the Spanish CIS administered (CIS publishes this questionnaire in Spanish only); resp 1 is ‘very interesting’ and 5 ‘not at all interesting’, so higher values mean less interest, and the IRW processing script drops ‘don’t know’/‘no answer’ (11-20 respondents per item, 92 responses in all) to missing.
spain_2023_science_benefits	The English in the _translated columns was machine-translated by IRW from the Spanish CIS administered (CIS publishes this questionnaire in Spanish only); resp 1 is ‘strongly agree’ and 5 ‘do not agree at all’ that the advance benefits society, so higher values mean less agreement, and the IRW processing script drops ‘don’t know’/‘no answer’ (53-111 respondents per item, 308 responses in all) to missing.
spain_2023_science_cuts	The English in the _translated columns was machine-translated by IRW from the Spanish CIS administered (CIS publishes this questionnaire in Spanish only), and the IRW processing script drops ‘don’t know’/‘no answer’ to missing (57-145 respondents per item, 261 responses in all; the private-companies item p16 loses the most, 145).
spain_2023_science_dangers	The English in the _translated columns was machine-translated by IRW from the Spanish CIS administered (CIS publishes this questionnaire in Spanish only); these four items repeat the item wording of the benefits battery (P.10) and differ only in the question, which asks about dangers; resp 1 is ‘strongly agree’ and 5 ‘do not agree at all’, and the IRW processing script drops ‘don’t know’/‘no answer’ (58-118 respondents per item, 337 responses in all) to missing.
spain_2023_science_engagement	The English in the _translated columns was machine-translated by IRW from the Spanish CIS administered (CIS publishes this questionnaire in Spanish only); resp 1 is ‘very often’ and 5 ‘never’, so higher values mean less frequent engagement; the question text reproduces a stray inverted question mark from the CIS questionnaire, and the IRW processing script drops ‘don’t know’/‘no answer’ (31 responses in all) to missing.
spain_2023_science_functions	The English in the _translated columns was machine-translated by IRW from the Spanish CIS administered (CIS publishes this questionnaire in Spanish only); resp 1 is ‘a lot’ and 5 ‘not at all’, so higher values mean science is seen as serving the function less, and the IRW processing script drops ‘don’t know’/‘no answer’ (11-41 respondents per item, 134 responses in all) to missing.
spain_2023_science_future	The English in the _translated columns was machine-translated by IRW from the Spanish CIS administered (CIS publishes this questionnaire in Spanish only); resp runs 1 = ‘many’ to 5 = ‘none’ with no 3, because the IRW processing script recodes the volunteered, not-read midpoint ‘Regular’ (code 3) to missing along with ‘don’t know’/‘no answer’ (83 responses on the benefits item, 121 on the risks item).
spain_2023_science_interest	The English in the _translated columns was machine-translated by IRW from the Spanish CIS administered (CIS publishes this questionnaire in Spanish only); resp 1 is ‘very interested’ and 5 ‘not at all interested’, so higher values mean less interest; the general-interest question (p1) reads out only ‘very, quite, not very or not at all’ although its answer list also offers ‘somewhat’ (3), as printed by CIS; and the IRW processing script drops ‘don’t know’/‘no answer’ (305 responses in all) to missing.
spain_2023_science_science	The English in the _translated columns was machine-translated by IRW from the Spanish CIS administered (CIS publishes this questionnaire in Spanish only); only the endpoints of the 1-5 scale are labelled (1 ‘does not associate it at all’, 5 ‘associates it a lot’), so resp 2-4 carry no option text, and the IRW processing script drops ‘don’t know’/‘no answer’ (6-139 respondents per item, 241 responses in all) to missing.
spain_2023_science_situations	The English in the _translated columns was machine-translated by IRW from the Spanish CIS administered (CIS publishes this questionnaire in Spanish only); only the scale endpoints are labelled (1 = ‘not at all useful’, 5 = ‘very useful’, so higher values mean more useful, the reverse of several sibling spain_2023_science tables), and the IRW processing script drops ‘not applicable’ (152 responses, mostly respondents with no children or relatives on the family-education item) and ‘don’t know’/‘no answer’ (91) to missing.
spain_2023_science_technology	The English in the _translated columns was machine-translated by IRW from the Spanish CIS administered (CIS publishes this questionnaire in Spanish only); only the endpoints of the 1-5 scale are labelled (1 ‘does not associate it at all’, 5 ‘associates it a lot’), so resp 2-4 carry no option text, and the IRW processing script drops ‘don’t know’/‘no answer’ (212 responses in all) to missing.
spain_2023_science_utility	The English in the _translated columns was machine-translated by IRW from the Spanish CIS administered (CIS publishes this questionnaire in Spanish only); only the scale endpoints are labelled (1 = ‘not at all useful’, 5 = ‘very useful’, so higher values mean more useful, the reverse of several sibling spain_2023_science tables), and the IRW processing script drops ‘don’t know’/‘no answer’ (27 responses) to missing.
spain_2024_ideology_discomfort	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3480 in Spanish only; the administered Spanish is in the base fields. Note also: codes run 1 = very uncomfortable to 4 = not at all uncomfortable; respondents who volunteered that the situation did not apply to them (e.g. no coworkers), and don’t-know/no-answer, are stored as missing.
spain_2024_ideology_identity	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3480 in Spanish only; the administered Spanish is in the base fields. Note also: codes run 1 = identifies a lot to 5 = does not identify at all; the unprompted middle answer (3, ‘neither a lot nor a little’), ‘does not apply’ (e.g. no profession or religion), don’t-know and no-answer are stored as missing, so the codes skip 3.
spain_2024_ideology_leaders	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3480 in Spanish only; the administered Spanish is in the base fields. Note also: leaders 6-12 (Rufian, Nogueras, Aizpurua, Esteban, Rego, Valido, Catalan) were rated only by respondents in the leader’s own autonomous community, so their n is 48-518 against 3276-3876 for the five national leaders; ‘does not know the leader’, don’t-know and no-answer are stored as missing.
spain_2024_ideology_parties	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3480 in Spanish only; the administered Spanish is in the base fields. Note also: parties 6-12 (ERC, Junts, EH Bildu, PNV, BNG, Coalicion Canaria, UPN) were rated only by respondents in the party’s own autonomous community, so their n is 61-533 against 3733-3875 for the five national parties; ‘does not know the party’, don’t-know and no-answer are stored as missing.
spain_2024_ideology_rights	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3480 in Spanish only; the administered Spanish is in the base fields. Note also: codes run 1 = strongly agree to 5 = strongly disagree; the unprompted middle answer (3, ‘neither agree nor disagree’), don’t-know and no-answer are stored as missing, so the codes skip 3.
spain_2024_ideology_society	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3480 in Spanish only; the administered Spanish is in the base fields. Note also: codes run 1 = strongly agree to 5 = strongly disagree; the unprompted middle answer (3, ‘neither agree nor disagree’), don’t-know and no-answer are stored as missing, so the codes skip 3.
spain_2024_ideology_sympathy	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3480 in Spanish only; the administered Spanish is in the base fields. Note also: ratings are of sympathy toward each party’s voters, not the party itself; only the endpoints 1 (‘no sympathy’) and 10 (‘a lot of sympathy’) are labelled; ‘does not understand the scale’, ‘does not know the party’, ‘cannot place it’ and no-answer are stored as missing.
spain_2024_ideology_territorial	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3480 in Spanish only; the administered Spanish is in the base fields. Note also: the printed CIS questionnaire lists only four identity options (it omits ‘more [regional] than Spanish’ and numbers ‘as Spanish as [regional]’ 4), but the data and the file’s value labels use five codes, 1 only Spanish … 5 only [regional], and the option text follows the value labels; ‘[regional]’ stands for the demonym of the respondent’s own autonomous community; ‘none of the above’, don’t-know and no-answer are stored as missing.
spain_2024_ideology_welfare	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3480 in Spanish only; the administered Spanish is in the base fields. Note also: the volunteered ‘neither agree nor disagree’ answer (code 3, not read aloud) is dropped from the response table, so codes run 1 strongly agree, 2 agree, 4 disagree, 5 strongly disagree.
spain_2024_politics_actions	The English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only.
spain_2024_politics_associations	The English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only.
spain_2024_politics_beliefs	The English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only.
spain_2024_politics_difficulty	The English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only; respondents could also volunteer ‘Ni fácil ni difícil’ (neither easy nor difficult; CIS code 3, not read aloud, about 4% per item), which the IRW processing script treats as missing along with ‘N.S.’ and ‘N.C.’, so the response values are 1, 2, 4 and 5 with no option row for 3.
spain_2024_politics_inperson	The English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only.
spain_2024_politics_online	The English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only; respondents who said the question did not apply because they had no devices or internet access (CIS code 7, about 1.5-1.7% per item) are treated as missing by the IRW processing script along with ‘N.S.’ and ‘N.C.’, so they have no response value or option row.
spain_2024_politics_political	The English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only; only the scale end points carry verbal labels (1 and 10 for ideology, 0 and 10 for vote probability), and the vote-probability question was not put to respondents without Spanish nationality (CIS code 97, about 4%), who are absent from that item.
spain_2024_politics_values	The English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only; respondents could also volunteer ‘Ni de acuerdo ni en desacuerdo’ (neither agree nor disagree; CIS code 3, not read aloud, under 0.5% per item), which the IRW processing script treats as missing along with ‘N.S.’ and ‘N.C.’, so the response values are 1, 2, 4 and 5 with no option row for 3.
spain_2024_values_difficulty	The English in the _translated columns was produced by IRW (machine translation), not by CIS. Respondents could also volunteer ‘Ni fácil ni difícil’ (neither easy nor difficult; code 3, not read aloud, 5.0% on p10 and 3.8% on p11), which the IRW processing script treats as missing, so the data hold only codes 1, 2, 4 and 5.
spain_2024_values_institutions	The English in the _translated columns was produced by IRW (machine translation), not by CIS. The volunteered middle response ‘Ni de acuerdo ni en desacuerdo’ (code 3, not read aloud) is treated as missing by the IRW processing script, so the stored scale runs 1 (strongly agree), 2, 4, 5 (strongly disagree) with no 3.
spain_2024_values_meaning	The English in the _translated columns was produced by IRW (machine translation), not by CIS. The volunteered middle response ‘Ni de acuerdo ni en desacuerdo’ (code 3, not read aloud; 8-34 respondents per item) is treated as missing by the IRW processing script, so the stored scale runs 1 (strongly agree), 2, 4, 5 (strongly disagree) with no 3.
spain_2024_values_political	The English in the _translated columns was produced by IRW (machine translation), not by CIS. Only the scale endpoints are labelled. probvoto was asked only of respondents with Spanish nationality; the 205 foreign nationals coded ‘No procede’ (96) are treated as missing.
spain_2024_values_values	The English in the _translated columns was produced by IRW (machine translation), not by CIS. The volunteered middle response ‘Ni de acuerdo ni en desacuerdo’ (code 3, not read aloud; 13, 9 and 16 respondents on p8_1, p8_2, p8_3) is treated as missing, so the stored scale runs 1 (strongly agree), 2, 4, 5 (strongly disagree) with no 3.
spain_2025_ageism_agreement	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3493 in Spanish only; the administered Spanish is in the base fields. Note also that the response values run 1, 2, 4, 5 with no option row for 3: respondents could volunteer a middle ‘Regular (ni de acuerdo ni en desacuerdo)’ answer (CIS code 3, not read aloud) which the IRW processing script drops to missing along with 8 ‘N.S.’ and 9 ‘N.C.’.
spain_2025_ageism_bureaucracy	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3493 in Spanish only; the administered Spanish is in the base fields. Two things to know about the responses: the P.11 block was asked only of respondents aged 65 and over (1,030 of the study’s 5,006 interviews), and the scale runs from 1 ‘En muchas ocasiones’ (most frequent) to 4 ‘Nunca’, with the not-read-aloud category 6 ‘No he realizado nunca esta gestion’ plus ‘N.S.’/‘N.C.’ dropped to missing by the IRW processing script, so those have no option row.
spain_2025_ageism_comparison	The English item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3493 in Spanish only; the administered Spanish is in the base columns. Note also that each of the three questions allowed a volunteered, not-read-aloud ‘both equally’ answer (CIS code 3, worded differently in each question) which the IRW processing script drops to missing along with 8 ‘N.S.’ and 9 ‘N.C.’, so the live scale is the two read-aloud categories only and no option row exists for those codes.
spain_2025_ageism_difficulty	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3493 in Spanish only; the administered Spanish is in the base fields. Note also that P.12 was asked only of respondents aged 65 and over (the questionnaire routes anyone under 65 past the block), and that the response values run 1, 2, 4, 5 with no option row for 3 or 6: respondents could volunteer a middle ‘Regular (ni facil ni dificil)’ answer (CIS code 3) or ‘nunca ha necesitado realizar esta gestion’ (code 6), neither of which the interviewer read aloud, and the IRW processing script drops both to missing along with 8 ‘N.S.’ and 9 ‘N.C.’.
spain_2025_ageism_elderpriority	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3493 in Spanish only; the administered Spanish is in the base fields. Note also that the 0-10 priority scale was labelled at its two ends only – 0 ‘Ninguna prioridad’ and 10 ‘Maxima prioridad’ – so values 1 to 9 carry no option text because the study printed none, and ‘N.S.’/‘N.C.’ (CIS codes 98 and 99) are dropped to missing by the IRW processing script.
spain_2025_ageism_elderslights	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3493 in Spanish only; the administered Spanish is in the base fields. Two things to know about the responses: the P.13 block was asked only of respondents aged 65 and over (1,030 of the study’s 5,006 interviews), and the fourth response category was coded 7 ‘Ninguna vez’ by CIS and recoded to 4 by the IRW processing script to give a contiguous 1-4 scale, so resp runs from 1 ‘Muchas veces’ (most frequent) to 4 ‘Ninguna vez’ (never).
spain_2025_ageism_problems	The English item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3493 in Spanish only; the administered Spanish is in the base columns. Two things to know about the responses: the scale runs from 1 ‘Muchos problemas’ (most problems) to 5 ‘Ningun problema’ (none), so a higher value means fewer perceived problems; and the values 1, 2, 4, 5 are non-contiguous because respondents could volunteer a middle ‘Ni muchos ni pocos’ answer (CIS code 3) that the interviewer did not read aloud, which the IRW processing script drops to missing along with ‘N.S.’ and ‘N.C.’, so no option row exists for it. Note also that p2’s administered wording is elliptical – the interviewer asked only ‘And young people under 35?’ immediately after p1, so p2’s response frame is the one stated in p1’s question.
spain_2025_ageism_youthpriority	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3493 in Spanish only; the administered Spanish is in the base fields. Note also that the 0-10 priority scale was labelled at its two ends only – 0 ‘Ninguna prioridad’ and 10 ‘Maxima prioridad’ – so values 1 to 9 carry no option text because the study printed none, and ‘N.S.’/‘N.C.’ (CIS codes 98 and 99) are dropped to missing by the IRW processing script.
spain_2025_ageism_youthslights	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3493 in Spanish only; the administered Spanish is in the base fields, and item 1’s diminutives ‘nino/a’ and ‘nene/a’ have no exact English equivalent. Two things to know about the responses: the P.16 block was asked only of respondents aged 18 to 34 (975 of the study’s 5,006 interviews), and the resp scale is deliberately non-contiguous – CIS coded the volunteered ‘(NO LEER) Regular’ category 3 and the IRW processing script drops it, while recoding CIS’s ‘Ninguna vez’ from 7 to 5, so resp runs 1 ‘Muchas veces’ (most frequent), 2 ‘Bastantes veces’, 4 ‘Pocas veces’, 5 ‘Ninguna vez’ (never) with no value 3.
spain_2025_ai_aspects	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3495 in Spanish only; the administered Spanish is in the base fields. Codes run 1 ‘Más beneficios’ to 3 ‘Más perjuicios’; the question was asked only of respondents who had heard of artificial intelligence, and the not-asked, ‘N.S.’ and ‘N.C.’ codes are dropped to missing by the IRW processing script.
spain_2025_ai_association	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3495 in Spanish only; the administered Spanish is in the base fields. Despite the study’s title, the question asks how much the respondent associates each word with technological advances in general, not with artificial intelligence; the 1-10 scale was labelled at its ends only (1 ‘En ninguna medida’, 10 ‘En gran medida’), so resp 2-9 carry no option text.
spain_2025_ai_awareness	The English item, prompt and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3495 in Spanish only; the administered Spanish is in the base fields. The table holds two questions: p10 (whether the respondent has heard of AI tools, asked of everyone and answered ‘Sí’ by 97%) and p18_1_1-p18_1_4 (whether they have heard of four recent AI events), which were asked only of those who had heard of AI tools, so the 124 respondents who answered ‘No’ at p10 have no event rows.
spain_2025_ai_comfort	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3495 in Spanish only; the administered Spanish is in the base fields. The table holds two different questions on the same 1-10 format: p13_1-p13_3 rate comfort with three situations performed by AI (1 ‘Totalmente incómodo/a’, 10 ‘Totalmente cómodo/a’) and p19 is self-rated knowledge of AI (1 ‘Ningún conocimiento’, 10 ‘Máximo conocimiento’). Only the endpoints are labelled, and all items were asked only of respondents who had heard of artificial intelligence; not-asked, ‘N.S.’ and ‘N.C.’ codes are dropped to missing by the IRW processing script.
spain_2025_ai_concern	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3495 in Spanish only; the administered Spanish is in the base fields. The items are about the use of personal data on the Internet, not AI; p6 is the elliptical follow-up ‘¿Y por parte de instituciones públicas?’ to p5 and inherits p5’s framing and scale. The volunteered ‘(NO LEER) Regular’ middle answer and ‘No usa Internet’ are dropped to missing by the IRW processing script along with ‘N.S.’/‘N.C.’, so resp runs 1, 2, 4, 5 with no option row for 3.
spain_2025_ai_datause	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3495 in Spanish only; the administered Spanish is in the base fields. The statements concern data collection on the Internet, not AI. Codes run 1 ‘Muy de acuerdo’ to 5 ‘Muy en desacuerdo’; the middle answer 3 ‘Ni de acuerdo ni en desacuerdo’ was not read aloud and is dropped to missing by the IRW processing script along with ‘N.S.’/‘N.C.’, so resp runs 1, 2, 4, 5 with no option row for 3.
spain_2025_ai_eventsagree	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3495 in Spanish only; the administered Spanish is in the base fields. Each event’s agreement question was asked only of respondents who had heard of AI tools and of that event. Codes run 1 ‘Muy de acuerdo’ to 5 ‘Muy en desacuerdo’; the middle answer 3 was not read aloud and is dropped to missing by the IRW processing script along with the not-asked, ‘N.S.’ and ‘N.C.’ codes, so resp runs 1, 2, 4, 5 with no option row for 3.
spain_2025_ai_frequency	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3495 in Spanish only; the administered Spanish is in the base fields. Codes run 1 ‘A diario’ (daily) to 5 ‘Una vez’ (once), so a higher value means less frequent use; each tool’s frequency was asked only of respondents who said they had used that tool in the last twelve months (143 for Perplexity, 1,987 for ChatGPT), and for p12_2_5 (other AI tool) interviewers were told to record the frequency of the main tool when the respondent used several. Tool names are in capitals as printed in the questionnaire (‘CHAT GPT’).
spain_2025_ai_impact	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3495 in Spanish only; the administered Spanish is in the base fields. The two items are unrelated questions that share only a response scale: p1 asks how much technological advances are changing people’s lives and p4 how important the privacy of one’s data on the Internet is (they correlate 0.14); codes run 1 ‘Mucho’ to 4 ‘Nada’, so a low value means more.
spain_2025_ai_possibilities	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3495 in Spanish only; the administered Spanish is in the base fields. Each statement completes the lead-in ‘La inteligencia artificial…’, which is in the instructions. The question was asked only of respondents who had heard of artificial intelligence; codes run 1 ‘Muy de acuerdo’ to 5 ‘Muy en desacuerdo’, and the middle answer 3 was not read aloud and is dropped to missing by the IRW processing script along with the not-asked, ‘N.S.’ and ‘N.C.’ codes, so resp runs 1, 2, 4, 5 with no option row for 3.
spain_2025_ai_regulation	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3495 in Spanish only; the administered Spanish is in the base fields. The question was asked only of respondents who had heard of artificial intelligence; codes run 1 ‘Muy de acuerdo’ to 5 ‘Muy en desacuerdo’, and the middle answer 3 was not read aloud and is dropped to missing by the IRW processing script along with the not-asked, ‘N.S.’ and ‘N.C.’ codes, so resp runs 1, 2, 4, 5 with no option row for 3.
spain_2025_ai_risk	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3495 in Spanish only; the administered Spanish is in the base fields. The items are about giving personal data on the Internet, not AI: each completes the stem ‘Cuando alguien proporciona datos personales en Internet, ¿cree Ud. que es muy probable, bastante, poco o nada probable que…?’. Codes run 1 ‘Muy probable’ to 4 ‘Nada probable’, so a low value means higher perceived risk.
spain_2025_ai_sectors	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3495 in Spanish only; the administered Spanish is in the base fields. Codes run 1 ‘Más beneficios’ to 3 ‘Más perjuicios’; the question was asked only of respondents who had heard of artificial intelligence, and the not-asked, ‘N.S.’ and ‘N.C.’ codes are dropped to missing by the IRW processing script. The Spanish ‘La seguridad’ can mean either security or safety.
spain_2025_ai_tools	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3495 in Spanish only; the administered Spanish is in the base fields. Codes are 1 ‘Sí’ (used the tool in the last twelve months), 2 ‘No’; the question was asked only of respondents who had heard of artificial intelligence, and the not-asked, ‘N.S.’ and ‘N.C.’ codes are dropped to missing by the IRW processing script. Tool names are in capitals as printed in the questionnaire (‘CHAT GPT’).
spain_2025_democracy_efficacy	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes this questionnaire only in Spanish; the administered Spanish is in the base fields. Note also that respondents could volunteer a middle ‘Ni de acuerdo ni en desacuerdo’ answer (CIS code 3, not read aloud, chosen by 0.1-0.7% of the sample) which the IRW processing script treats as missing, so the response values run 1, 2, 4, 5 with no option row for 3.
spain_2025_democracy_internal	The item and response wording for this CIS survey question is in Spanish, as administered; the English in the _translated columns was produced by IRW, not by CIS.
spain_2025_democracy_judiciary	The response scale respondents heard was 1 Muy de acuerdo / 2 De acuerdo / 4 En desacuerdo / 5 Muy en desacuerdo, with a fifth, not-read-aloud category 3 ‘Ni de acuerdo ni en desacuerdo’ plus 8 ‘N.S.’ and 9 ‘N.C.’ that the IRW processing script drops to missing, which is why resp jumps from 2 to 4 and no option row exists for 3; the English in the _translated columns was produced by IRW, not by CIS.
spain_2025_democracy_media	The response scale respondents heard was 1 Muy de acuerdo / 2 De acuerdo / 4 En desacuerdo / 5 Muy en desacuerdo, with a fifth, not-read-aloud category 3 ‘Ni de acuerdo ni en desacuerdo’ plus 8 ‘N.S.’ and 9 ‘N.C.’ that the IRW processing script drops to missing, which is why resp jumps from 2 to 4 and no option row exists for 3; the English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only.
spain_2025_democracy_parties	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3497 in Spanish only; the administered Spanish is in the base fields. Note also that the response values run 1, 2, 4, 5 with no option row for 3: respondents could volunteer a middle ‘Ni de acuerdo ni en desacuerdo’ answer (CIS code 3, not read aloud) which the IRW processing script drops to missing along with 8 ‘N.S.’ and 9 ‘N.C.’.
spain_2025_democracy_priorities	The item, instruction and response-option wording in the _translated columns is IRW’s own English; CIS publishes this questionnaire only in Spanish, and the administered Spanish is in the base fields.
spain_2025_democracy_system	The English item, option and question text for this table was machine-translated by IRW from the Spanish that CIS actually administered in Estudio 3497; the administered Spanish is the authoritative text and sits in the base columns.
spain_2025_democracy_trust	Respondents answered on an 0-10 trust scale on which only the endpoints were labelled (‘Ninguna confianza’ at 0, ‘Total confianza’ at 10), so option_text is deliberately blank for resp 1-9 rather than repeating the numeral; the codes 98 ‘N.S.’ and 99 ‘N.C.’ are dropped to missing by the IRW processing script and have no option row; and the English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only.
spain_2025_europe_agreement	The English item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3523 in Spanish only; the administered Spanish is in the base fields. The scale runs 1 ‘Muy de acuerdo’, 2 ‘Bastante de acuerdo’, 4 ‘Poco de acuerdo’, 5 ‘Nada de acuerdo’, so a lower value means more agreement; resp 3 ‘Ni de acuerdo ni en desacuerdo’ sits between 2 and 4 but was not read out and was recorded only when volunteered (6 to 15 respondents per item). The three items are separate tax questions rather than a scale, and ‘N.S.’/‘N.C.’ answers are dropped to missing by the IRW processing script.
spain_2025_europe_attention	The English item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3523 in Spanish only; the administered Spanish is in the base fields. Both items run from 1 (most interested, most informed) to 4 (‘Nada’, ‘Nada informado/a’), so a lower value means more interest or information, and the few ‘N.S.’/‘N.C.’ answers are dropped to missing by the IRW processing script.
spain_2025_europe_awareness	The English instruction, item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3523 in Spanish only; the administered Spanish is in the base fields, and each item is a bare institution name completing the stem. resp 1 is ‘Sí, ha oído hablar’ and 2 ‘No, no ha oído hablar’, so a higher value means less awareness; the few ‘N.C.’ answers are dropped to missing by the IRW processing script. The table is unweighted, so its shares differ from CIS’s published, weighted results (for example 7.5% had not heard of the Council of the EU here against 14.1% published).
spain_2025_europe_effects	The English instruction, item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3523 in Spanish only; the administered Spanish is in the base fields, and each item is a short phrase completing the stem. The seven items were read in a rotated order, so p8_1 to p8_7 is not the order every respondent heard. p8_5 is shipped as ‘desfavorecidas’, the spelling in the CIS data file, where the questionnaire misprints it ‘defavorecidas’. resp runs 1 ‘Más bien beneficiosa’, 2 ‘Ni beneficiosa ni perjudicial’, 3 ‘Más bien perjudicial’, and the middle answer was not read out and was recorded only when volunteered. ‘N.S.’/‘N.C.’ answers are dropped to missing by the IRW processing script (7.1% of respondents on p8_1, salaries), and the table is unweighted, so its shares differ from CIS’s published, weighted results.
spain_2025_europe_impact	The English item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3523 in Spanish only; the administered Spanish is in the base fields. The table mixes two scales: p4 and p7 run 1 ‘Mucho’ to 4 ‘Nada’, while p5 and p6 run 1 ‘Más bien positivo’, 2 ‘Ni positivo ni negativo’, 3 ‘Más bien negativo’, with the middle answer not read out and recorded only when volunteered. A lower value means more impact, a more positive view or more satisfaction throughout, and p4 asks how much EU decisions affect Spaniards’ lives, not whether that is good or bad. ‘N.S.’/‘N.C.’ answers are dropped to missing by the IRW processing script.
spain_2025_europe_policy	The English instruction, item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3523 in Spanish only; the administered Spanish is in the base fields. The five measures were read in a rotated order, so p12_1 to p12_5 is not the order every respondent heard. p12_5’s Spanish keeps the questionnaire’s misprint ‘económicamenta’ (for ‘económicamente’); the English renders the intended word. resp is 1 ‘A favor’, 2 ‘En contra’, and ‘N.S.’/‘N.C.’ answers (1.6% to 2.9% per item) are dropped to missing by the IRW processing script.
spain_2025_europe_spending	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3523 in Spanish only; the administered Spanish is in the base fields. Codes run 1 ‘Demasiados’ (too many public resources) to 3 ‘Demasiado pocos’ (too few), so a higher value means the service is judged more underfunded; ‘N.S.’ and ‘N.C.’ answers (0.5-3.7% per item) are dropped to missing by the IRW processing script, and the table is unweighted, so its shares differ from the weighted marginals CIS publishes.
spain_2025_europe_trust	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3523 in Spanish only; the administered Spanish is in the base fields. Note also that the 1-10 trust scale was labelled at its ends only (1 ‘Ninguna confianza’, 10 ‘Total confianza’), so resp 2-9 carry no option text, and that each institution was asked only of respondents who said at the previous question that they had heard of it; the not-asked code (N.P., up to 14.8% of respondents for the European Council) and ‘N.S.’/‘N.C.’ are dropped to missing by the IRW processing script, so missingness on each item partly reflects unfamiliarity with the institution.
spain_2025_fears_leaders	The English instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3534 in Spanish only; the administered Spanish is in the base fields, and the items themselves are the leaders’ names. Note also that each leader was rated only by respondents who said they knew him or her, and that the 1-10 scale was labelled at its two ends only (1 ‘Muy mal’, 10 ‘Muy bien’), so values 2 to 9 carry no option text.
spain_2025_fears_prospect	The English item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3534 in Spanish only; the administered Spanish is in the base fields. Note also that on both items the middle option (resp 3: ‘Satisfecho/a’ on p10, ‘Igual, ni mejor ni peor’ on p11) was not read out to respondents and was recorded only when volunteered.
spain_2025_inequality_inequality	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3522 in Spanish only; the administered Spanish is in the base fields. Note also that the 1-10 scale was labelled at its two ends only (1 ‘No existe ninguna desigualdad’, 10 ‘La desigualdad es máxima’), so values 2 to 9 carry no option text.
spain_2025_inequality_leaders	The English instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3522 in Spanish only; the administered Spanish is in the base fields, and the items themselves are the leaders’ names. Note also that each leader was rated only by respondents who said they knew him or her, and that the 1-10 scale was labelled at its two ends only (1 ‘Muy mal’, 10 ‘Muy bien’), so values 2 to 9 carry no option text.
spain_2025_inequality_personal	The English item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3522 in Spanish only; the administered Spanish is in the base fields. Note also that on p4 and p10 the middle option ‘Regular’ (resp 3) was not read out to respondents and was recorded only when volunteered, so it is rarely used (31 and 108 of about 4,000 answers), whereas p11’s middle option ‘Las mismas’ was read out.
spain_2025_inequality_recognition	The English item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3522 in Spanish only; the administered Spanish is in the base fields, and the items are the names of four party leaders, left unchanged in English. The IRW processing script recodes the responses: CIS coded 1 ‘Conoce’ and 7 ‘No conoce’, and the table stores ‘No conoce’ as resp 0, while 9 ‘N.C.’ (3 respondents per leader) is dropped to missing. Recognition is near ceiling, 97.8% to 99.7% on every leader.
spain_2025_inequality_trajectory	The English item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3522 in Spanish only; the administered Spanish is in the base fields. The IRW processing script reorders the responses, so resp does not follow the CIS codes (on which 1 is ‘Más’): resp 1 = less/decrease (CIS code 2), 2 = the same (CIS code 3), 3 = more/increase (CIS code 1). On p5 and p6 the middle option ‘Igual que ahora’ was not read out and was recorded only when volunteered; p5 looks back ten years although that option is printed as ‘Igual que ahora’. ‘N.S.’/‘N.C.’ answers are dropped to missing, which removes 457 of 4,004 answers on p6. The three items are separate questions about Spain in the past, Spain in the future and the autonomous communities in the future, not a scale.
spain_2025_sanitation_aicomfort	Respondents rated each situation on a 1-10 comfort scale on which only the endpoints were labelled (‘Totalmente incomodo/a’ at 1, ‘Totalmente comodo/a’ at 10), so option_text is deliberately blank for resp 2-9; N.S./N.C. answers are dropped to missing by the IRW processing script; and the English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only.
spain_2025_sanitation_aiconcern	The English item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3531 in Spanish only; P24 was asked as the follow-up ‘¿Y por parte de instituciones públicas?’ to P23’s question about private companies, so it asks about concern over public institutions using one’s clinical data to develop health AI; resp 3 ‘Regular’ was a volunteered middle answer that interviewers did not read out (17 and 18 respondents).
spain_2025_sanitation_airegulation	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3531 in Spanish only; the administered Spanish is in the base fields. Response 3 ‘Ni de acuerdo ni en desacuerdo’ was not read aloud by the interviewer (the question offers four agreement categories) and was volunteered by only 9, 4 and 9 respondents on the three items; ‘N.S.’/‘N.C.’ answers were dropped to missing by the IRW processing script.
spain_2025_sanitation_coordination	The English item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3531 in Spanish only; P13’s resp 3 ‘No le han recetado medicamentos en la sanidad pública’ (55 respondents) is a not-read-aloud not-applicable code that interviewers marked for people not prescribed medicines in the public system (including private-only users), so P13 is nominal rather than a 1-3 ordinal scale; P11’s ‘Otra respuesta’, ‘N.S.’ and ‘N.C.’ answers were dropped to missing by the IRW processing script.
spain_2025_sanitation_diagnostics	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3531 in Spanish only; per the interviewer instruction, ultrasounds for pregnancy and tests done as part of cancer screening programmes were to be recorded as No, and N.S./N.C. answers were dropped to missing by the IRW processing script.
spain_2025_sanitation_digital	The English item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3531 in Spanish only; the administered Spanish is in the base fields. Volunteered answers that interviewers did not read out (‘No utiliza la sanidad pública’, ‘No usa recetas de la sanidad pública’, ‘No sabe qué es’, and on P15 ‘La persona dice que no es cierto que se puedan consultar’) were dropped to missing by the IRW processing script along with N.S./N.C., so each item’s resp covers only the read-out options.
spain_2025_sanitation_hospital	The English item, prompt and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3531 in Spanish only; these questions were asked only of the 249 respondents admitted to a public hospital in the previous twelve months, the five P9C satisfaction rows were read in rotating order, points 2-9 of the P9C 1-10 scale carry no label, and N.S./N.C. answers were dropped to missing by the IRW processing script.
spain_2025_sanitation_mental	The English item, prompt and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3531 in Spanish only; all items were asked only of respondents who needed a mental-health consultation in the past year and were seen in the public health system, and the four P8D yes/no items only of those treated mainly by a psychiatrist, psychologist or other specialist rather than the family doctor (n=155 vs 239 for P8E), while N.S./No recuerda and N.C. answers were dropped to missing by the IRW processing script.
spain_2025_sanitation_primary	The English item, prompt and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3531 in Spanish only; these questions were asked only of respondents who had consulted a public-system family doctor in the past twelve months, P5D_3 (waiting more than an hour) only of those seen in person at the surgery, and P5H only of those who had a telephone consultation; N.S./N.C. answers were dropped to missing by the IRW processing script (notably 71 ‘don’t know’ answers on P5E_2, the nursing professional).
spain_2025_sanitation_regional	The English item, prompt and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3531 in Spanish only; ‘don’t know’ answers, which were common on these questions (469, 421, 452 and 628 on P19_1 to P19_4, the most on financing), were dropped to missing by the IRW processing script along with refusals.
spain_2025_sanitation_services	The English instructions, item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3531 in Spanish only; points 2-9 of the 1-10 scale carry no label, and N.S./N.C. answers were dropped to missing by the IRW processing script, which removes 535 of 2427 answers (22%) on P4ES_4, the 061/112 emergency lines, against at most 133 on any other item.
spain_2025_sanitation_specialist	The English item, prompt and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3531 in Spanish only; these questions were asked only of respondents who had seen a public-system medical specialist in the past twelve months; on P7E, resp 6 is not a point beyond ‘very poor’ but a volunteered, not-read-aloud answer ‘it depends: better in some consultations, worse in others’ (8 respondents), which the IRW processing script kept, so treat it as non-ordinal; N.S./N.C. answers were dropped to missing.
spain_2025_sanitation_system	The English question and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3531 in Spanish only; the three items are not coded in a common direction – P1 runs 1 = works quite well to 4 = works badly, P2 runs 1 = very dissatisfied to 10 = very satisfied, and P12 is coded 1 = improved, 2 = worsened, 3 = stayed the same, so its codes are not in order of meaning; don’t-know and no-answer responses were dropped to missing by the IRW processing script.
spain_2025_sex_attitudes	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3515 in Spanish only; the administered Spanish is in the base fields. Note also that the middle answer (code 3, ‘Ni de acuerdo ni en desacuerdo’) was not read out to respondents and was recorded only when volunteered (18-106 respondents per item), and that the order of the five statements was rotated between interviews.
spain_2025_sex_condom	The English item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3515 in Spanish only; the administered Spanish is in the base fields. Two things to know about the responses: according to a footnote in the CIS questionnaire, a programming error meant that women who reported sexual experience with both men and women were never asked these questions; and the not-read-aloud answer ‘Solo ha tenido relaciones sexuales orales’ (code 7) and ‘N.C.’ were dropped to missing by the IRW processing script, so they have no option row.
spain_2025_sex_frequency	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3515 in Spanish only; the administered Spanish is in the base fields. Note also that items b2_3 to b2_6 were asked only of respondents who reported sexual experience with other people, so their counts are lower than those of b2_1 and b2_2.
spain_2025_sex_identity	The English item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3515 in Spanish only; the administered Spanish is in the base fields. The table bundles four separate questions with different response scales: a7, a9 and a13 are 1 = yes, 2 = no, while a10 runs from 1 ‘Muy satisfecho’ to 5 ‘Nada satisfecho’, so resp = 1 means yes on three items and very satisfied on one. a9 was asked only of the roughly 115 respondents reporting a non-normative gender identity, and a10’s code 3 (‘Ni satisfecho ni insatisfecho’) was not read out and was recorded only when volunteered.
spain_2025_tourism_importance	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3521 in Spanish only; the administered Spanish is in the base fields. Note also that the volunteered, not-read-aloud middle answer ‘(NO LEER) Regular’ (code 3) is dropped to missing by the IRW processing script along with 8 ‘N.S.’ and 9 ‘N.C.’, so the table carries only resp 1, 2, 4 and 5.
spain_2026_confidence_acquisition	The English in the _translated columns was produced by IRW (machine translation), not by CIS. resp 1 merges two administered answers, ‘acquired it myself’ and ‘another household member acquired it’ (respondents could mark both), which the IRW processing script combines into one ‘yes’ code; ‘don’t know’ and ‘no answer’ (32 responses across the four goods) are treated as missing.
spain_2026_confidence_expectations	The English in the _translated columns was produced by IRW (machine translation), not by CIS. p10’s item text includes the interviewer’s printed definition of ‘bienes duraderos’ (cars, furniture, appliances, computers; never housing); ‘don’t know’ and ‘no answer’ are treated as missing.
spain_2026_confidence_macro	The English in the _translated columns was produced by IRW (machine translation), not by CIS. p15’s question quotes the prior-year inflation figure (3.2%) that was read to respondents in this May 2026 wave.
spain_2026_confidence_prospect	The English in the _translated columns was produced by IRW (machine translation), not by CIS. resp 2 ‘Igual’ (the same) was not read to respondents: each question offered only better or worse, and ‘the same’ was recorded only when volunteered (396-1,162 responses per item), so the midpoint of this 3-point scale was never offered.
spain_2026_confidence_retrospect	The English in the _translated columns was produced by IRW (machine translation), not by CIS. Interviewers read only ‘better or worse’, so resp 2 ‘Igual’ (the same) is a volunteered answer, chosen by 20-47% of respondents per item; ‘don’t know’ and ‘no answer’ are treated as missing.
spain_2026_disinformation_concern	The English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only; ‘N.S.’ (don’t know) and ‘N.C.’ (no answer) responses were set to missing when the table was built.
spain_2026_disinformation_hostility	The English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only; ‘N.S.’ (don’t know) and ‘N.C.’ (no answer) responses were set to missing when the table was built.
spain_2026_disinformation_humorclimate	The English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only; for p30, resp 3 (Igual, the same) was not read out to respondents and was recorded only when volunteered, and N.S. (don’t know) and N.C. (no answer) responses were set to missing when the table was built.
spain_2026_disinformation_humorlimits	The English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only; the unprompted middle answers (P.13 “Depende”, P.24 “No estoy seguro/a”, code 3) and “N.S.”/“N.C.” responses were set to missing when the table was built, so p24 takes the values 1, 2 and 4 only.
spain_2026_disinformation_humorperception	The English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only; ‘N.S.’ (don’t know) and ‘N.C.’ (no answer) responses were set to missing when the table was built.
spain_2026_disinformation_memes	The English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only; p14 keeps the CIS codes, so its last option (Nunca o rara vez, never or rarely) is resp 7 rather than 5; the p15 items were asked only of respondents with p14 = 1-4 (those who receive memes at least every few months); and N.S. (don’t know) and N.C. (no answer) responses were set to missing when the table was built.
spain_2026_disinformation_offense	The English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only; N.S. (don’t know) and N.C. (no answer) responses were set to missing when the table was built.
spain_2026_disinformation_personalhumor	The English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only; resp keeps CIS’s own codes, so the ‘Nada’/‘Nunca’ endpoint of p17, p20 and p25 is stored as 7 (not 5 or 4), and resp 3 on p21 is the volunteered, not-read-out ‘Ni de acuerdo ni en desacuerdo’, while ‘N.S.’/‘N.C.’ answers were set to missing when the table was built.
spain_2026_disinformation_regulation	The English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only; p3 and p6 also offered a volunteered, not-read-out third answer (p3 ‘Depende’, it depends; p6 ‘yes, but only voluntarily’), which was set to missing along with N.S. (don’t know) and N.C. (no answer) when the table was built, so every item is stored as 1 = Sí (yes), 2 = No.
spain_2026_elections_campaign	The English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only; on the interest question (p2) resp 3 ‘Regular’ (so-so) was not read to respondents – the question offered only a great deal, quite a lot, little or no interest – and was recorded only when volunteered (78 of 4032 respondents).
spain_2026_elections_garcia	The English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only; respondents who said at the preceding question that they did not know José Ignacio García (233 of 4032, CIS code 0 ‘No procede’) were not asked this question and are absent from the table, and only the scale end points (1 ‘Poco’, 10 ‘Mucho’) carry verbal labels.
spain_2026_elections_gavira	The English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only; respondents who said at the preceding question that they did not know Manuel Gavira (206 of 4032, CIS code 0 ‘No procede’) were not asked this question and are absent from the table, and only the scale end points (1 ‘Poco’, 10 ‘Mucho’) carry verbal labels.
spain_2026_elections_leaderscampaign	The English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only; respondents who said they did not know a candidate (CIS code 97 ‘No conoce’, printed as a column of the grid: 20, 25, 206, 233 and 199 of 4032 for the five candidates in item order) are absent from that candidate’s item, and only the scale end points (1 ‘Muy mal’, 10 ‘Muy bien’) carry verbal labels.
spain_2026_elections_maillo	The English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only; respondents who said at the preceding question that they did not know Antonio Maíllo (199 of 4032, CIS code 0 ‘No procede’) were not asked this question and are absent from the table, and only the scale end points (1 ‘Poco’, 10 ‘Mucho’) carry verbal labels.
spain_2026_elections_montero	The English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only; respondents who said at the preceding question that they did not know María Jesús Montero (25 of 4032, CIS code 0 ‘No procede’) were not asked this question and are absent from the table, and only the scale end points (1 ‘Poco’, 10 ‘Mucho’) carry verbal labels.
spain_2026_elections_moreno	The English in the _translated columns was produced by IRW, not by CIS, which publishes this questionnaire in Spanish only; respondents who said at the preceding question that they did not know Juan Manuel Moreno (20 of 4032, CIS code 0 ‘No procede’) were not asked this question and are absent from the table, and only the scale end points (1 ‘Poco’, 10 ‘Mucho’) carry verbal labels.
spain_2026_international_confidence	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3564 in Spanish only; the administered Spanish is in the base fields, and the English gives the organisations’ standard English names and acronyms (OTAN as NATO, FMI as IMF, CPI as ICC). The 1-10 confidence scale was labelled only at its ends, 1 ‘Ninguna confianza’ and 10 ‘Máxima confianza’, so values 2 to 9 carry no option text because the study printed none. ‘N.S.’/‘N.C.’ answers (CIS codes 98 and 99) are dropped to missing by the IRW processing script, which removes up to 8% of respondents for the less familiar bodies (489 of 6,001 for the International Criminal Court, 455 for the OECD, 398 for the WTO).
spain_2026_international_eu	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3564 in Spanish only; the administered Spanish is in the base fields. ‘N.S.’/‘N.C.’ answers are dropped to missing by the IRW processing script.
spain_2026_international_influence	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3564 in Spanish only; the administered Spanish is in the base fields, and the English gives the institutions’ standard English names and acronyms (ONU as UN, CPI as ICC). The 1-10 influence scale was labelled only at its ends, 1 ‘Ninguna influencia’ and 10 ‘Máxima influencia’, so values 2 to 9 carry no option text because the study printed none. ‘N.S.’/‘N.C.’ answers (CIS codes 98 and 99) are dropped to missing by the IRW processing script, which removes up to 7% of respondents for the less familiar bodies (441 of 6,001 for the International Criminal Court, 414 for the OECD, 393 for the WTO), so the number answering varies from item to item.
spain_2026_international_media	The English item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3564 in Spanish only; the administered Spanish is in the base fields, and the English renders ‘Nada’ as ‘None’ on p7 (interest) and ‘Not at all’ on p9 and p10. p10’s ‘estos conflictos’ refers back to the conflicts named in the preceding questions (Ukraine, Gaza, the Strait of Hormuz) and is shipped as worded, not expanded. The scale runs 1 ‘Mucho’ to 4 ‘Nada’, and ‘N.S.’/‘N.C.’ answers are dropped to missing by the IRW processing script (328 of 6,001 on p10).
spain_2026_international_threat	The English item and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3564 in Spanish only; the administered Spanish is in the base fields. p16’s ‘estos conflictos’ refers back to a preceding question, not in this table, that asked for the greatest current international risk (great-power escalation, Ukraine, the Middle East, humanitarian and migration crises, nuclear weapons). Both items keep the CIS coding 1 ‘Mucho’ to 4 ‘Nada’, so a higher resp means less affected or less worried; ‘N.S.’/‘N.C.’ answers are dropped to missing by the IRW processing script.
spain_2026_international_un	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3564 in Spanish only; the administered Spanish is in the base fields, and the Spanish instruction keeps the questionnaire’s stray ‘afirmaciones:.’ punctuation. ‘N.S.’/‘N.C.’ answers are dropped to missing by the IRW processing script, which removes 277 of 6,001 respondents (4.6%) on onu_5, the Security Council veto item.
spain_2026_love_apps	The English in the _translated columns was machine-translated by IRW from the Spanish CIS administered (CIS publishes this questionnaire in Spanish only); resp 1 is ‘strongly agree’ and 5 ‘do not agree at all’, so higher values mean less agreement; the question was asked only of respondents who had heard of dating apps; and the IRW processing script drops the volunteered, not-read-out middle answer ‘Regular’ (3; 97 responses) as well as ‘don’t know’/‘no answer’ (1,409) to missing, so resp takes only the values 1, 2, 4 and 5.
spain_2026_love_association	The English in the _translated columns was machine-translated by IRW from the Spanish CIS administered (CIS publishes this questionnaire in Spanish only); resp 1 is ‘a lot’ and 5 ‘not at all’, so higher values mean a weaker association with love; and the IRW processing script drops the volunteered, not-read-out middle answer ‘Regular’ (3; 143 responses) as well as ‘don’t know’/‘no answer’ (250) to missing, so resp takes only the values 1, 2, 4 and 5.
spain_2026_love_attitudes	The English in the _translated columns was machine-translated by IRW from the Spanish CIS administered (CIS publishes this questionnaire in Spanish only); resp 1 is ‘strongly agree’ and 5 ‘do not agree at all’, so higher values mean less agreement; and the IRW processing script drops the volunteered, not-read-out middle answer ‘Regular’ (3; 143 responses) as well as ‘don’t know’/‘no answer’ (376) to missing, so resp takes only the values 1, 2, 4 and 5.
spain_2026_love_conceptions	The English in the _translated columns was machine-translated by IRW from the Spanish CIS administered (CIS publishes this questionnaire in Spanish only); resp 1 is ‘Sí’ (yes) and 2 is ‘No’, and the IRW processing script drops ‘don’t know’/‘no answer’ (232 responses) to missing.
spain_2026_love_desire	The English in the _translated columns was machine-translated by IRW from the Spanish CIS administered (CIS publishes this questionnaire in Spanish only); the two items were asked of different subsets – p19 only of respondents with no current romantic relationship (1,278) and p22a only of respondents not currently in love (1,381) – so the table covers 1,508 of the survey’s 5,007 respondents; resp 1 is ‘a lot’ and 5 ‘not at all’, so higher values mean less desire; and the IRW processing script drops the volunteered, not-read-out middle answer ‘Regular’ (3; 31 responses) as well as ‘don’t know’/‘no answer’ (30) to missing, so resp takes only the values 1, 2, 4 and 5.
spain_2026_love_gestures	The English in the _translated columns was machine-translated by IRW from the Spanish CIS administered (CIS publishes this questionnaire in Spanish only); resp 1 is ‘yes’ (important) and 2 ‘no’, and ‘don’t know’/‘no answer’ (53 responses) are missing.
spain_2026_love_importance	The English in the _translated columns was machine-translated by IRW from the Spanish CIS administered (CIS publishes this questionnaire in Spanish only); resp 1 is ‘very important’ and 4 ‘not at all important’, so higher values mean less importance; and the IRW processing script drops ‘don’t know’/‘no answer’ (289 responses) to missing.
spain_2026_love_usage	The English in the _translated columns was machine-translated by IRW from the Spanish CIS administered (CIS publishes this questionnaire in Spanish only); the items are a filtered sequence – p15 was asked of everyone, p15a only of the 4,609 respondents who had heard of dating apps, and p15b only of the 1,349 who had created a profile – so later items cover smaller subsets; resp 1 is ‘yes’ and 2 ‘no’, and ‘don’t know’/‘no answer’ (6 responses) are missing.
spain_2026_prostitution_pornography	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3525 in Spanish only; the administered Spanish is in the base fields. Note also that the response values run 1, 2, 4, 5 with no option row for 3: respondents could volunteer a middle ‘Ni de acuerdo ni en desacuerdo’ answer (CIS code 3, not read aloud, 28-63 respondents per item) which the IRW processing script drops to missing along with 8 ‘N.S.’ and 9 ‘N.C.’. The order of the statements was rotated between interviews.
spain_2026_prostitution_prostitution	The English item, instruction and response-option wording for this table was produced by IRW, not by CIS, which publishes Estudio 3525 in Spanish only; the administered Spanish is in the base fields. Note also that the response values run 1, 2, 4, 5 (1 = ‘Muy de acuerdo’, 5 = ‘Nada de acuerdo’) with no option row for 3: respondents could volunteer a middle ‘Ni de acuerdo ni en desacuerdo’ answer (CIS code 3, not read aloud) which the IRW processing script drops to missing along with 8 ‘N.S.’ and 9 ‘N.C.’. The order of the statements was rotated between interviews.
spelling_assessment_study1	Words were presented aurally (each spoken three or four times) and typed by participants; item_text is the target word, and instructions paraphrase the procedure as the paper reports it rather than the on-screen wording.
spelling_assessment_study2	Words were presented aurally and typed by participants; item_text is the target word, and instructions paraphrase the procedure as the paper reports it for Study 1, whose word set Study 2 reused. The deposit (and so this table) holds 72 of the 109 words the authors appear to have administered.
SPQVS_Barnby_2017_OEQ	The source paper calls this the seven-item OEQ7, but the study’s data and this table carry six items; the paper’s reported OEQ7 total (M 9.90, SD 2.84) is the sum of these six.
srivani_2022_education4	The questionnaire offered five options (Strongly agree 2 to Strongly disagree -2), but the response table omits the 20 ‘Disagree’ (-1) responses recorded in the source file for items ELN, LET, ELS, TDE and ECT, so only the Neutral, Agree and Strongly agree options appear here
stachl_2020_belonging	Each item was a cartoon illustration; item_text transcribes only the text in its speech and thought bubbles, and the six negatively-worded (Low SB) items are stored reverse-scored, so for them resp 4 means Do Not Relate
stenhouse_2026_math_anxiety	The response table drops the endpoint responses of the two 1-10 anxiety ratings (items 17.1 and 20.1): the source stored 1 and 10 as the labels ‘1 - Not anxious’ and ‘10 - Very anxious’, which the processing script discarded, so 54 and 40 responses respectively are missing and those items only take values 2-9
stolz_2015_death_attitudes	The item text and response options are the administered German. The English item text in item_text_translated was produced by IRW; the study published no English for these six items. The English response options are the study’s own, taken from its English rendering of the same four answer categories used elsewhere in the survey.
strohacker_2024_arms_readiness	The study administered these items in English to its US cohort and in an unpublished German translation to its German cohort, and the data carry no cohort flag, so item_text is not the wording German-cohort respondents read
strojny_2026_gmi	The English translations (item_text_translated, section_prompt_translated) were written by IRW from the Polish wording respondents read; they are not the GMI’s original English item set, which the deposit does not include.
su_2024_gad7	The item text is the administered Chinese as the study recorded it, which is a different Chinese translation from the publisher’s official one. The study published no instructions or response-option labels, so those base fields are blank; the English in the _translated columns, including the anchors (0 = Not at all .. 3 = Nearly every day), is the publisher’s official English GAD-7, not a translation of what respondents saw, and items 5 and 7 differ in phrasing from the Chinese.
su_2024_phq9	The item text is the administered Chinese as the study recorded it, which is a different Chinese translation from the publisher’s official one (all nine stems differ). The study published no instructions or response-option labels, so those base fields are blank; the English in the _translated columns, including the anchors (0 = Not at all .. 3 = Nearly every day), is the publisher’s official English PHQ-9, not a translation of what respondents saw, and some stems differ in phrasing (item 3’s Chinese reads roughly ‘difficulty falling asleep, restless sleep or sleeping too much’).
sugiura_2015_disaster_characteristics	The questionnaire was administered in Japanese; the item text shown is the authors’ own English translation from the study’s data file, which differs in minor wording from the versions printed in the paper’s Table 2 for 8 items
sumner_2022_ipip_neo	The study did not publish its form, so the item wording is the published IPIP-NEO-120 text, assigned to the study’s item numbers by the instrument’s standard order. Facet membership and keying were confirmed against the study’s own facet scores, but the order of items that share both a facet and a keying direction could not be confirmed. Responses are stored raw, so reverse-keyed items are not reversed.
sun_2016_risky_choice_graph	Respondents saw each gamble pair as a coordinate graph with a caption (the paper prints it only for pair 1: ‘A denotes 80% chance of receiving ¥50 and 20% chance of receiving 0; B denotes 40% chance of receiving ¥100 and 60% chance of receiving 0’), so item_text is blank and option_text gives each option’s win probability and prize in the paper’s Table 1 shorthand; graph-scale condition is in cov_graph_condition
sun_2021_blockchain_loan_adoption	The survey of SMEs was run in China and does not state its administration language; only the authors’ English questionnaire is published, so item_text is that English rather than the wording respondents read. Only the 1 and 7 anchors are labelled, and the order of items within each three-item construct rests on the paper’s numbering, which no published item-level statistic confirms
sun_2024_innovation	The survey of 281 Chinese SME CEOs does not state its administration language and publishes only the authors’ English wording, which is what is shipped here, so it may not be the wording respondents read
sun_2024_performance_expectancy	The survey of 281 Chinese SME CEOs does not state its administration language and publishes only the authors’ English wording, which is what is shipped here, so it may not be the wording respondents read
sun_2024_price_value	The survey of 281 Chinese SME CEOs does not state its administration language and publishes only the authors’ English wording, which is what is shipped here, so it may not be the wording respondents read
sun_2024_risk	The survey of 281 Chinese SME CEOs does not state its administration language and publishes only the authors’ English wording, which is what is shipped here, so it may not be the wording respondents read
sun_2024_task_tech_fit	The survey of 281 Chinese SME CEOs does not state its administration language and publishes only the authors’ English wording, which is what is shipped here, so it may not be the wording respondents read
sun_2024_usage	The survey of 281 Chinese SME CEOs does not state its administration language and publishes only the authors’ English wording, which is what is shipped here, so it may not be the wording respondents read
sun_2025_morality_study2_colleagueclose	The item text here is the administered Chinese, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors.
sun_2025_morality_study2_colleagueknow	The item text here is the administered Chinese, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors.
sun_2025_morality_study2_colleaguelike	The item text here is the administered Chinese, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors.
sun_2025_morality_study2_informant-moral-character	The item text here is the administered Chinese, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors.
sun_2025_morality_study2_nemotion	The item text here is the administered Chinese, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors.
sun_2025_morality_study2_pemotion	The item text here is the administered Chinese, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors.
sun_2025_morality_study2_self-moral-character	The item text here is the administered Chinese, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. The section prompt is already English in the published table, so its translated column stays empty.
sun_2026_eap_content	The questionnaire was distributed through a Chinese survey platform (Wenjuanxing) to students and teachers at a Chinese art institute, but the study publishes only an English version, so the item text shown is the authors’ English and may not be the wording respondents read; the paper does not state the administered language
sun_2026_eap_difficulty	The questionnaire was distributed through a Chinese survey platform to students and teachers at a Chinese art institute, but the study publishes only an English version, so the item text shown is the authors’ English and may not be the wording respondents read; the paper does not state the administered language. The question asks respondents to rate each difficulty by its severity, but the published response scale is an agreement scale (Strongly Disagree to Strongly Agree)
sun_2026_eap_methods	The questionnaire was distributed through a Chinese survey platform to students and teachers at a Chinese art institute, but the study publishes only an English version, so the item text shown is the authors’ English and may not be the wording respondents read; the paper does not state the administered language. The question asks respondents to rate each teaching method by its effectiveness, but the published response scale is an agreement scale (Strongly Disagree to Strongly Agree)
sun_2026_eap_usage	The questionnaire was distributed through a Chinese survey platform to students and teachers at a Chinese art institute, but the study publishes only an English version, so the item text shown is the authors’ English and may not be the wording respondents read; the paper does not state the administered language. The question asks respondents to rate the importance or frequency of each scenario, but the published response scale is an importance scale (Not Important at All to Very Important)
sun_2026_tiktok_travel	The survey was administered in Chinese but only the authors’ English questionnaire is published, which is shipped here; that English form prints the disagree-side anchors as 2 = ‘Somewhat disagree’ and 3 = ‘Disagree’, an order that is almost certainly transposed but is reproduced as printed because no source confirms the administered labels
sv-maia2_randelovic_2021_delta	The item text here is the administered Serbian, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors.
sv-maia2_randelovic_2021_erq	IRW table uses a 1-5 scale (the paper’s ERQ is normally 1-7); the itemtext anchor labels for the 1-5 version are inferred from the 1-7 wording, not sourced from the study’s own materials
sv-maia2_randelovic_2021_maia	The item text here is the administered Serbian, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. option_text mixes languages within one field, carrying Serbian nikad and English Always as the two endpoint anchors of the same scale. Always is mapped to itself so the translated column is whole rather than half-filled.
szameitat_2015_multitask_examples	Item text is the study’s English questionnaire; respondents in Turkey, the Netherlands and Germany (about 60% of this table) answered the authors’ own Turkish, Dutch and German translations, which were not published
szameitat_2015_occupation_multitask	The questionnaire was also administered in German, Dutch and Turkish translations (209 of the 347 respondents here, per cov_country), whose wording was never published, so the English version shipped is what only the UK, US and International respondents read
takacs_2026_aces	As for takacs_2026_phq9: the codebook’s English wording, not the Hungarian wording respondents read.
takacs_2026_gad7	As for takacs_2026_phq9: the codebook’s English wording, not the Hungarian wording respondents read.
takacs_2026_maas	As for takacs_2026_phq9: the codebook’s English wording, not the Hungarian wording respondents read.
takacs_2026_phq9	The item text is the English wording in the deposit’s codebook, the authors’ rendering of a questionnaire given in Hungarian to Hungarian-speaking adults; the Hungarian wording respondents read is not published.
talaifar_2025_study2_thermometer	The prompt respondents read says “We would like to get your feelings toward both Democrats and Republicans”, but the two items ask for ratings of conservatives and liberals; the wording is reproduced as administered. Only the Cold, Neutral and Warm points of the 0-100 thermometer are labelled.
tan_2026_stem_technology_integration	Response options 1-5 carry no labels in the source, so the direction of the scale is undocumented; negatively worded items 9 and 10 correlate positively (r = 0.22 to 0.48) with the other 13 items, and the source does not say whether they were reverse-coded
tangarife_2026_ai_benefits	As for tangarife_2026_ai_use: IRW-written English from the Spanish statements.
tangarife_2026_ai_reservations	As for tangarife_2026_ai_use: IRW-written English from the Spanish statements.
tangarife_2026_ai_use	The English translations (*_translated) were written by IRW from the Spanish statements respondents read; they are not the study’s own.
tangarife_2026_ease_of_use	As for tangarife_2026_ai_use: IRW-written English from the Spanish statements.
tangarife_2026_research_barriers	As for tangarife_2026_ai_use: IRW-written English from the Spanish statements.
tangarife_2026_social_influence	As for tangarife_2026_ai_use: IRW-written English from the Spanish statements.
tangarife_2026_use_intention	As for tangarife_2026_ai_use: IRW-written English from the Spanish statements.
taylorabdulai_2025_incentives	Item text is the source paper’s Table 4 labels for the question ‘What will make you accept COVID-19 vaccine’ (the questionnaire itself was not published); the pattern of answers (120 of 377 respondents endorse none of the seven factors, 179 exactly one) indicates a select-all-that-apply question, so resp = 0 (‘No’ in the deposit) most likely means the factor was not selected rather than explicitly rejected
tears	The wording shown is the paper’s description of each rating dimension, not the text observers saw on screen, which is not published. Four items are 0-100 scales (negative valence, positive valence, arousal, genuineness); the other seven are emotion percentages that had to sum to 100 across the seven, so they are not independent ratings. id is the video clip being rated, not a person; the observer is in rater.
teicher_2015_mace_distress_helpless	Item text is the published MACE-X form, which differs from the version administered for four items (Attempt_sex_sib, Intercourse_sib, H_Adults_argue, O_Adults_argue – the study’s own paper describes the last two as hearing and observing adults arguing), so those four ship without item text; also, resp = 0 (Helpless box not checked) is recorded for every respondent who never experienced the event, which is 92% of the zeros
teicher_2015_mace_distress_terrified	Item text comes from the published MACE-X form, which words four positions differently from the version this study administered, so item_text is blank for H_Adults_argue and O_Adults_argue (hearing vs. observing adults argue) and Attempt_sex_sib and Intercourse_sib (witnessing attempted vs. actual intercourse with a sibling); resp = 1 means the respondent ticked ‘Terrified’ for that event and resp = 0 means they did not, which includes the large majority of rows (26,542 of 30,503) from respondents who reported never experiencing the event
teicher_2015_mace_emot_neglect	Responses are stored as answered (1 = Yes, 0 = No) and are not reverse-scored: for the three positively worded items (felt loved, felt important or special, family a source of strength and support) a 1 indicates the absence of emotional neglect
teicher_2015_mace_nonverbal	This table holds 5 of the 6 items the MACE scores as parental non-verbal emotional abuse: item 65 of the MACE-X, ‘You felt that you had to shoulder adult responsibilities.’ (33.9% Yes in the source paper), is in the study’s data file but was not included in the response table
teicher_2015_mace_phys_neglect	Responses are stored as answered (1 = Yes, 0 = No) and are not reverse-scored, so for the three positively worded items (P_protect, P_ER, Looked_out_each_other) a 1 indicates the absence of physical neglect; item_text for P_ER is the study’s experimental MACE-X wording (‘… if the need ever arose.’), while the authors’ final MACE adds ‘, or would have if needed.’, and which wording the online administration used is not published
teicher_2015_mace_sexual	The source paper’s Table 9 labels the item scored from o_touch_them as ‘Other adults touched or fondled you in sexual way’, but the study’s published MACE and MACE-X forms word it ‘Had you touch their body in a sexual way’, which is the text shipped here; the response pattern supports the forms (o_fondled 10.4% Yes vs o_touch_them 5.8%, mirroring the parental items Fondled 2.9% vs Touch_them 1.5%)
teicher_2015_mace_witness_parent	Item text comes from the published MACE-X form, whose items 31/32 (adults arguing with your mother / father) differ from the two arguing items this study administered – the paper describes them as hearing and observing adults arguing, and the data agree (448 of 458 who observed also heard) – so H_Adults_argue and O_Adults_argue ship without item text; Adults_hit_med is the father item (adults hurt your father so that he needed medical attention) despite lacking the _f suffix
teicher_2015_mace_witness_sib	Item text comes from the published MACE-X form, which at two positions words the items differently from the version this study administered (the study’s paper describes them as separate attempted and actual intercourse items), so Attempt_sex_sib and Intercourse_sib ship without item text
temesgen_2025_elephant_conserv_attitude	Respondents were interviewed in Tigrinya from a questionnaire the authors wrote in English, and no Tigrinya wording was published, so the item text is the authors’ English
temesgen_2025_elephant_park_attitude	Respondents were interviewed in Tigrinya from a questionnaire the authors wrote in English, and no Tigrinya wording was published, so the item text is the authors’ English; for item Q7 the response data do not reproduce the percentages the source paper reports for that statement (38.7% agree / 7.9% neutral / 53.4% disagree in the data vs 34.17 / 11.65 / 54.18 in the paper’s Table 6), while the other six items match to 0.01 points
teq_novak_2021_scbs	The Santa Clara Brief Compassion Scale was administered in Czech, but the Czech wording is not published, so the item text is English, not what respondents read. SCBCS_1-5 are assumed to follow the published item order; nothing in the deposit states it and the data cannot separate item 2 from item 3. The English is transcribed from a source that ends item 1 at “compassion” (other printings add “for him or her”) and prints item 5 with the typo “when the seem to be in need”. The anchors are this study’s 7-point ones (1 = Not at all true of me, 7 = Very true of me), not the source’s 5-point format.
teq_novak_2021_selfesteem	Item text for this Czech-administered Rosenberg Self-Esteem Scale table is the English RSES wording, not the Czech wording respondents read, and the item codes carry no labels: they are matched to the commonly circulated RSES order (negatively worded items 2, 5, 6, 8, 9), which the data support for each item’s polarity and for item 8 but which does not distinguish item 3 from 4, item 1 from 10, or the negatively worded items 2, 5, 6, 9 from one another. Responses are stored so that a higher value means lower self-esteem on every item (0 = Strongly Agree for positively worded items, 0 = Strongly Disagree for negatively worded ones), the opposite of the source paper’s statement that higher scores indicate higher self-esteem
teq_novak_2021_spirit	The Daily Spiritual Experience Scale was administered in a 15-item Czech version whose wording is not published, so the item text is Underwood’s English original, not what respondents read; the Czech form drops Underwood’s item 5, and codes are matched through the published renumbering. Direction: 1 = Many times a day and 6 = Never or almost never, so a higher value means less frequent spiritual experience – the paper’s statement that a higher score means more spirituality contradicts its own anchors, and the data follow the anchors. SPIRIT_15 (the closeness-to-God item) is on four points running the same way, 1 closest, reversed relative to Underwood’s printed order. Misspelt codes such as SIPRIT_8 and SPRIT_14 are source typos kept as join keys.
test_taking_much_2025_ef	Item EF05 (‘I could have tried harder on the previous test block’) appears to be stored already reverse-scored: it correlates +0.22 to +0.34 with the other effort items and the paper’s reliability matches it as stored, so its response labels are shipped reversed (1 = strongly agree, 4 = strongly disagree) relative to the paper’s stated 1 = strongly disagree scale; the deposit’s EF_01/EF_02 scale scores reverse it a second time.
thanh_2025_attitude	Respondents answered a Vietnamese translation that was not published, so the item text is the authors’ English; the responses stored for ATT4 and ATT5 (items the authors dropped from their analysis) are statistically indistinguishable from uniform random numbers – near-flat counts across 1-5 and correlations of at most 0.11 with any other item in the study – so they should not be analysed as responses to that wording without checking with the authors
thanh_2025_environ_concern	The study was administered in Vietnamese but only the authors’ English wording is published, so that English is shown here; item EC6 (‘I worry about the greenhouse effect’) was dropped by the authors and its responses are essentially uncorrelated with every other item in the deposited data (r = 0.01-0.06 with EC1-EC5, including ‘I worry about global warming’), so treat it with caution
thanh_2025_green_behavior	The item text is the authors’ published English; the survey was administered in Vietnamese after translation by the authors’ bilingual experts, and the Vietnamese wording was not published. Items EGB6-EGB8 were dropped by the authors for low factor loadings, and in the deposited data their responses are spread almost evenly across 1-5 and are uncorrelated with every other item (
thanh_2025_green_knowledge	Respondents answered a Vietnamese translation that was not published, so the item text is the authors’ English; separately, responses to GK5 and GK6 (the two items the authors dropped) are spread almost evenly over 1-5 and are essentially uncorrelated with GK1-GK4 (
thanh_2025_perceived_control	Item wording is the study’s own English rendering of a questionnaire administered in Vietnamese, whose wording was not published
theobald_2017_group_dynamics	The dominator item’s wording is shipped as printed in the paper’s Table 1 (‘today’s [topic] activity’); the Methods text quotes it as ‘today’s group activity’, and the specific topic shown to students varied by survey wave (cov_topic)
thomeczek2025_les	The Hamburg LES was administered to German experts in German; the item text shown is the author’s English codebook, and the German wording is not in this deposit.
tian2026_digital_competence	The English item wording in item_text_translated is a machine translation made by IRW from the administered Chinese, not the authors’ own English; the response anchors are the paper’s English endpoint labels because the Chinese anchors were not published, and three typos in the source’s Chinese (items 14, 20, 23) are kept as administered
tomioka_2022_srh_future_needs	The English in the _translated columns is the study’s own rendering and condenses the Japanese options (e.g. B9-4 ‘Consultation with expert’ translates a longer option about establishing expert consultation and supervision systems); resp 1 means the option was circled
tomioka_2022_srh_importance	The study’s English rendering is shorter than the administered Japanese, which asks about explaining the effects of the cancer and its treatment on sexual and reproductive function (including that there are none); both questions offered a fourth option, ‘Not important’ (4), which no nurse chose, so it has no row here
tomioka_2022_srh_sufficiency	The item text is the administered Japanese, with the study’s own English alongside, and the two versions differ. The Japanese B6 question offered only four options, worded ‘in place’ (整っている) rather than ‘carried out’, while the English gives all four items uniform ‘Sufficient’ .. ‘Insufficient’ anchors and a fifth ‘I don’t know’ for B6. The Japanese B5 question has no ‘sufficiently’. The ‘I don’t know’ option offered for B3-B5 has no row because those respondents were excluded from the data.
tomioka_2022_srh_support_types	The English in the _translated columns is the study’s own rendering: its Q7 carries only the heading ‘Support for sexual and reproductive issues’ where the administered Japanese asks nurses to circle every kind of support they provide, and several options are condensed (e.g. B7-1 ‘Referral to other professionals’ for referral to specialist physicians, counsellors or treatment institutions); resp 1 means the option was circled, and B7-11 is the ‘None’ option
torices_2025_bapq	The BAPQ was administered to this sample in Spanish, but neither the data deposit nor a published paper contains the Spanish item wording, so item_text is the standard English BAPQ wording standing in for what respondents read; the response labels are the study’s own Spanish labels from its codebook, with English alongside. Response 6 (‘Muy a menudo’, very often) occurs only on BAPQ_01 (17 of 108 respondents) and never on the other 35 items, where answers pile up at 5, so the top category appears not to have been available or recorded for items 2-36. The 15 reverse-keyed items (1, 3, 7, 9, 12, 15, 16, 19, 21, 23, 25, 28, 30, 34, 36) are stored unreversed
torok_2025_ai_acceptance	The item text here is the administered Hungarian, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. These are author-written survey items with no canonical instrument behind them, so no published English exists to use instead. Typos in the source are left verbatim in the administered text.
torok_2025_data_disclosure	The item text here is the administered Hungarian, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. These are author-written survey items with no canonical instrument behind them, so no published English exists to use instead. Typos in the source are left verbatim in the administered text.
torok_2025_data_security	The item text here is the administered Hungarian, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. These are author-written survey items with no canonical instrument behind them, so no published English exists to use instead. Typos in the source are left verbatim in the administered text.
torok_2025_discourse_responsibility	The item text here is the administered Hungarian, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. These are author-written survey items with no canonical instrument behind them, so no published English exists to use instead. Typos in the source are left verbatim in the administered text.
torok_2025_facebook_uses	The item text here is the administered Hungarian, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. These are author-written survey items with no canonical instrument behind them, so no published English exists to use instead. Typos in the source are left verbatim in the administered text.
torok_2025_internet_use_frequency	The item text here is the administered Hungarian, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. These are author-written survey items with no canonical instrument behind them, so no published English exists to use instead. Typos in the source are left verbatim in the administered text.
torok_2025_legality_beliefs	The item text here is the administered Hungarian, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. These are author-written survey items with no canonical instrument behind them, so no published English exists to use instead. Typos in the source are left verbatim in the administered text.
torok_2025_manipulation_fear	The item text here is the administered Hungarian, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. These are author-written survey items with no canonical instrument behind them, so no published English exists to use instead. Typos in the source are left verbatim in the administered text.
torok_2025_news_consumption	The item text here is the administered Hungarian, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. These are author-written survey items with no canonical instrument behind them, so no published English exists to use instead. Typos in the source are left verbatim in the administered text.
torok_2025_news_source_frequency	The item text here is the administered Hungarian, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. These are author-written survey items with no canonical instrument behind them, so no published English exists to use instead. Typos in the source are left verbatim in the administered text.
torok_2025_news_source_trust	The item text here is the administered Hungarian, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. These are author-written survey items with no canonical instrument behind them, so no published English exists to use instead. Typos in the source are left verbatim in the administered text.
torok_2025_social_media_effects	The item text here is the administered Hungarian, which is what respondents read. The English in the _translated columns was produced by IRW rather than by the study’s authors. These are author-written survey items with no canonical instrument behind them, so no published English exists to use instead. Typos in the source are left verbatim in the administered text.
tran_2023_gad7	The GAD-7 was administered in Vietnamese to medical students in Vietnam, but the study’s deposit and paper publish no item wording, so the item and option text shipped is the instrument publisher’s official English rather than what respondents read; the mapping of GAD1..GAD7 to the canonical items rests on the column numbering and is only partly confirmed by the data
tran_2023_phq9	The PHQ-9 was administered in Vietnamese, but no Vietnamese wording is published in the paper or its data deposit, so the item and option text shipped here is the instrument publisher’s official English rather than the sentences respondents read
trivia_fastrich_2017	Five of the 244 trivia questions were reworded part way through data collection, so the item text IRW ships is the later of two administered wordings: item 135 and 234 read ‘poisonous’ rather than ‘venomous’ for the first 387 and 266 responses, item 167 read ‘what county’ for its first 370, item 224 read ‘gas marks’ for its first 330, and item 267’s ‘Hippolyte Mege-Mouries’ appeared for 322 responses with its accented characters corrupted; both wordings of each are visible in the response table’s own cov_question column.
trusz_2025_nfi	The item and option text is the author’s English rendering of a Polish administration; the Polish wording survives only as truncated data-file column names. The respondents’ midpoint read ‘Nie mam zdania’ (‘I have no opinion’) rather than the English ‘neither yes nor no’. Typos in the English appendix are kept as printed. Six half-point answers in the source file (on items 11, 23, 24, 46 and 69) were truncated to whole numbers by the IRW processing script.
tsai_2017_treeit_tam_bi	The TAM questionnaire was administered to Mandarin-Chinese-reading older adults in Taiwan, but only the study’s English version is published, so the item text is that English rendering
tsai_2017_treeit_tam_peou	The TAM questionnaire was administered to Mandarin-Chinese-reading older adults in Taiwan, but only the study’s English version is published, so the item text is that English rendering
tsai_2017_treeit_tam_pu	The TAM questionnaire was administered to Mandarin-Chinese-reading older adults in Taiwan, but only the study’s English version is published, so the item text is that English rendering
tuason_2021_covid_coping_enjoy	Each respondent selected exactly five options. The source file’s own 0/1 selection columns were filled in only for the low and high well-being groups the paper analysed, so IRW builds the items from the source’s list of each respondent’s picks: every respondent has their five selections coded 1 and the rest 0
turner_2022_sr_belief_change	The study’s S2 File numbers its items by factor-loading rank rather than by the question codes in the data, so the text-to-code mapping was reconstructed by reproducing the paper’s published loadings and cross-loadings. Q34_SRchange and Q35_Srchange publish near-identical loadings (.693/.036 and .690/.037) and are separated only by their endorsement means, so a swap of those two texts cannot be fully excluded.
turpochaparro_2026_self_esteem	The item text is the administered Spanish. The English in item_text_translated was produced by IRW. The deposit does not label the 1-4 response scale or name the instrument, so option text and instructions are blank.
twod_rotation_mather2023	The 304 items are figure stimuli, so item_text is blank; each picture is R2Dfigures/fig.png in the study’s OSF “Item Stimuli” archive, numbered as in the item code. Participants chose among eight picture options (A-C and E-G, D “None of these.”, H “I don’t know.”), but resp is scored correctness, so option_text carries Incorrect/Correct scoring labels rather than the options presented.
ucla_BrummerHoffman_2021	The second item is coded ucla_left_out, but the Portuguese actually administered reads ‘Abandonado’ (abandoned) rather than a translation of the canonical ‘How often do you feel left out?’; item_text_translated follows the administered Portuguese (‘Abandoned’), while the study’s own English questionnaire prints the canonical wording. The other two item translations and the anchors come from that English questionnaire, but the English of the shared stem was translated by IRW.
uffler_2017_lecture_seating	Item text is shown in English. Uffler et al. administered the MSLQ in French translation and neither the paper nor the deposit publishes the French wording, so the English printed in the study’s own codebook ships instead of what respondents read.
uffler_2017_seat_reasons	The seven items are categories the authors assigned to free-text answers to one open question (“Why did you sit in that place today?”), not a printed checklist; resp 1 means the reason was mentioned. The category labels are the study’s French from its S1 figure, but the question itself ships only in the paper’s English rendering because the French original is not published.
useche_2022_cbq	The item text is the English root questionnaire from the deposit. Respondents outside the UK and Australia read a translation into their country’s main language, which is not published.
useche_2022_rprs	As for useche_2022_cbq: the English root form; most respondents read a translation that is not published.
uti_newlands_2023_gad	The item wording is the GAD-7’s standard English form, not text from the study’s own materials: neither the paper nor the OSF codebook prints the items. resp is stored 1-4, one above the form’s printed 0-3 scoring, so the GAD-7 total is sum(resp) - 7. Codes are matched to wording by the form’s printed item numbering. The data pin gad_5 (restless) and support gad_4 (trouble relaxing), but nothing confirms the order of gad_1, gad_2, gad_3, gad_6 and gad_7.
uti_newlands_2023_phq	The item wording is the PHQ-9’s standard English form, not text from the study’s own materials: neither the paper nor the OSF codebook prints the items. Codes are matched to wording by the form’s printed item numbering. The data pin phq_3, phq_4, phq_8 and phq_9 and only narrowly support phq_2, but nothing confirms the order of phq_1, phq_5, phq_6 and phq_7.
uti_newlands_2023_wpai	The item wording is the WPAI:SHP V2.0 English template from its rights holder, Reilly Associates, not the study’s own form. The study adapted the questionnaire to urinary tract infection, but that wording is not published, so the template’s placeholder PROBLEM is left in place rather than guessed. wpai_2 to wpai_4 are hours entered as numbers, not rating scales, so they have no response options. Codes are matched to wording by the form’s question numbering; the skip pattern and the paper’s published impairment scores confirm it. The form tells respondents who worked 0 hours to skip wpai_5, but 12 baseline and 7 retest respondents answered it anyway; those responses are kept as recorded.
valdivia_2023_oms	Item text is the administered Mexican Spanish from the study’s S1 File. The English in item_text_translated is the instrument’s official OMS-HC-15 wording rather than a translation of the Spanish, and for item 2 the two diverge – the Spanish reads ‘comodo(a) trabajando con el’ (comfortable working with him) and appears to omit a verb, while the English reads ‘I would be as willing to work with him/her’. The English of the instructions was written by IRW, not by the study.
valverdeberrocoso_2021_tictip	The English item wording is the study’s own (paper Table 2), but its item 9 renders the administered ‘mensajeria instantanea (v.gr. Rayuela)’ as ‘(e.g. Scratch)’; IRW ships ‘Rayuela’ to match what respondents read. The English of the instructions was written by IRW, not by the study.
vanteffelen_2020_aq_hostility	The paper describes a 10-item hostility scale anchored ‘extremely uncharacteristic of me’ to ‘extremely characteristic of me’, but the deposited data hold the 8 canonical Buss-Perry hostility items and the file’s own value labels are a Dutch agreement scale whose midpoint is ‘Weet niet’ (don’t know); the text shipped follows the deposit. The English in item_text_translated is Buss & Perry’s (1992) published wording, not a translation of the Dutch, and the English of the response options was written by IRW.
vanteffelen_2020_foa	The Forms of Aggression questionnaire was administered in Dutch and item_text ships that Dutch verbatim from the study’s own SPSS variable labels; the English in the _translated columns was produced by IRW, not by the authors. The questionnaire’s instructions are not published in either the paper or the deposit, so instructions is blank – the items are bare behaviour phrases (‘Ruzie beginnen te maken’) whose frequency framing survives only in the response anchors.
vaporova_2020_imitation	Item text for this table describes behaviour that was coded, not questions that were asked: the three items are the demonstrated target actions of a mitten-on-a-puppet imitation task scored 0/1 by observers, and the single instructions line is the paper’s English rendering of an utterance made to German-speaking infants and 4-year-olds.
VEI_Brazillian_Shiramizu_2018_EI	The items are the administered Brazilian Portuguese, but the response anchors (1 = Does not describe me well, 5 = Describes me very well) are the English wording the paper reports; the Portuguese anchors respondents saw were not published, and points 2-4 are unlabelled.
verbagg	Item text is shown in English: Vansteelandt’s questionnaire was administered in Dutch to Dutch-speaking Belgian students and no Dutch wording is published in the lme4 deposit or in De Boeck and Wilson’s online materials, so the English rendering reproduced in the psychotools and difR package documentation ships instead, with the ‘I would curse’ form of the do-mode items written as the parallel of the published ‘I would want to curse’; note also that resp=1 merges the administered ‘perhaps’ and ‘yes’ options because the table stores lme4’s dichotomised r2
Veterans_Affairs_SSVF_Survey_2016-17	Item 58’s wording is taken from the 2015 form (‘since you finished receiving services’), because the FY2016-2017 wording is not published; the neighbouring items 56 and 57 were reworded to ‘since you started’ by the FY2018-2020 form, so what FY2016-2017 respondents read for item 58 may differ. Item 7 (‘Q8 into’, stored value always 1) carries no text, and items 43, 49 and 55 – the write-in sixth ‘Other Supportive Services’ row – carry only the block stem with no service label. Item 3 runs 1 = Very Dissatisfied to 5 = Very Satisfied, confirmed against the published year-end totals and its +0.87 correlation with item 1; this is the reverse of the direction the VA data dictionary states for the FY2018-2020 satisfaction items.
vocabulary_iq	Only items 1-45 are vocabulary questions. Items 46-75 are the optional post-test personality survey (S1-S30), whose 1-5 agreement responses were lost when the processing script scored the whole table against the vocabulary answer key, so every one of those 30 items reads resp = 0 for every respondent and carries no response option text. Among the vocabulary items, resp = 0 means ‘not the correct pair’, which also absorbs the 0.87% of cells where the respondent skipped the question; the 13.3% of cells where they chose ‘don’t know’ are missing rather than 0.
VSOCFP_Sebo_2024	The item text here is the administered French SOC-13/SOC-8 wording from the study’s Multimedia Appendix 1. The English in the _translated columns is not the study’s own: the item wording is quoted from a published English SOC-13 item table (Islam et al. 2025) and the 1/7 anchors from a different published source (Bonanato et al. 2022), so it is a parallel English version rather than a translation of the French, and the English instructions were translated from the French by IRW.
wallace_2026_apply	The two items were worded differently in the two pooled studies (a consulting firm, MCM Consulting, in Study S1c, cov_study = 4; a volunteer organization, XYZ Organization, in Study S2c, cov_study = 7), so item_text gives both wordings labelled by study; responses are coded 1 = not at all to 7 = very much in both studies, Study S2c having been reverse-recoded from its printed order (itemcov_recoded = 1).
wallace_2026_belong	The items were worded differently in the pooled studies (about a class with a professor in Study 2, cov_study = 1; about an organization in Studies S1c and S2c, cov_study = 4 and 7), so item_text gives both wordings labelled by study; responses are coded 1 = not at all to 7 = extremely in all studies, Study S2c having been reverse-recoded from its printed order (itemcov_recoded = 1), and belong4R (‘alienated’) is not reverse-scored.
wallace_2026_cengage	The items were worded differently in the pooled studies (about a professor in Study 2, cov_study = 1; MCM Consulting in Study S1c, cov_study = 4; XYZ Organization in Study S2c, cov_study = 7), so item_text gives each study’s wording labelled by study, and cengage4 was asked in Study 2 only; responses run 1 = Strongly agree to 7 = Strongly disagree, so higher values mean less engagement.
wallace_2026_certain	Two designs share this table: Study 2 and Study S1c (cov_study = 1, 4) asked three global questions after all the intelligence-belief items, while Studies S1a, S1b, S2a, S2b (certain_item1-2) and S2c (certain1-3) asked How certain do you feel about the response that you just provided? right after each belief item (itemcov_design), so certain1-3 carry two wordings labelled by study; responses run 1 = low to 7 = high certainty in all studies, Study S2c having been reverse-recoded from its printed 1 = Very Certain order (itemcov_recoded = 1).
wallace_2026_climate	Responses run 1 = Strongly agree to 7 = Strongly disagree in all three studies (a higher value means less agreement); organized was asked in Study S2c (cov_study = 7) only.
wallace_2026_concern	Items concern2-5 were worded differently in the two pooled studies (about your boss in Study S1c, cov_study = 4; about others/other volunteers in Study S2c, cov_study = 7), so item_text gives both wordings labelled by study; responses are coded 1 = not at all to 7 = very much in both studies, Study S2c having been reverse-recoded from its printed order (itemcov_recoded = 1).
wallace_2026_fit	Items were worded for a different target in each pooled study (a professor in Study 2, cov_study = 1; MCM Consulting in Study S1c, cov_study = 4; XYZ Organization in Study S2c, cov_study = 7), so item_text gives all three wordings labelled by study; responses run 1 = Strongly agree to 7 = Strongly disagree, so a higher value means less perceived fit.
wallace_2026_interest	Items were worded for a different target in each pooled study (taking a class with a professor in Study 2, cov_study = 1; working at MCM Consulting in Study S1c, cov_study = 4), so item_text gives both wordings labelled by study; responses run 1 = not at all to 7 = extremely, the opposite direction from the agree-disagree wallace_2026_interest_org table.
wallace_2026_interest_org	The item was worded for a different organization by study – working at MCM Consulting in Studies S1a and S1b (cov_study = 2, 3), volunteering for the XYZ Organization in Study S2c (cov_study = 7) – and its wording in Studies S2a and S2b (cov_study = 5, 6; a volunteer tutoring organization) is not recorded in the source files, so item_text gives only the three documented wordings; responses run 1 = Strongly agree to 7 = Strongly disagree (higher = less interest), the opposite direction from wallace_2026_interest, with the S2a/S2b anchors unlabelled in the source and assumed to match.
wallace_2026_iqbelief	Responses run 1 = Strongly agree to 7 = Strongly disagree, so a higher value means disagreeing that intelligence is fixed (a more growth-oriented belief); the intermediate labels (Agree … Disagree) were shown only in Studies 2 and S1c (cov_study = 1, 4), and the other studies’ respondents saw numbers at points 2-6.
wallace_2026_learn	Items learn1-3 named a different target organization in the two pooled studies (MCM Consulting in Study S1c, cov_study = 4; XYZ Organization in Study S2c, cov_study = 7), so item_text gives both wordings labelled by study; responses are coded 1 = not at all to 7 = very much in both studies, Study S2c having been reverse-recoded from its printed order (itemcov_recoded = 1).
wallace_2026_mcm_check	The fixed/malleable and MCMmanip items name MCM Consulting in Studies S1a, S1b and S1c (cov_study = 2, 3, 4) and XYZ Organization in Studies S2a, S2b and S2c (cov_study = 5, 6, 7), and the IQB items ask about a professor (Study 2, cov_study = 1), so item_text gives each study’s wording labelled by study; responses run 1 = Strongly agree to 7 = Strongly disagree in every study, so a higher value means more disagreement with the statement.
wallace_2026_mcm_culture	Responses run 1 = Strongly agree to 7 = Strongly disagree, so a higher value means less agreement with the statement; the items refer to the fictitious organization MCM Consulting described to participants in Studies S1a and S1b (cov_study = 2, 3).
wallace_2026_rec	Items rec1-2 were worded differently in the two pooled studies (a friend in the consulting industry and MCM Consulting in Study S1c, cov_study = 4; a friend looking for a volunteer opportunity and XYZ Organization in Study S2c, cov_study = 7), so item_text gives both wordings labelled by study; responses are coded 1 = not at all to 7 = very much in both studies, Study S2c having been reverse-recoded from its printed order (itemcov_recoded = 1).
wallace_2026_reflection	Each rating was asked right after an intelligence-belief statement and refers back to it: the _item1 ratings follow You have a certain amount of intelligence, and you can’t really do much to change it. and the _item2 ratings follow Your intelligence is something about you that you can’t change very much., so the paired items share their wording.
wallace_2026_s2_engage	Responses run 1 = Strongly agree to 7 = Strongly disagree, so a higher value means less anticipated engagement in the professor’s course.
wallace_2026_s2_like	Responses run 1 = Strongly agree to 7 = Strongly disagree, so a higher value means less liking of the professor.
wallace_2026_s2_perf	Responses run 1 = Strongly agree to 7 = Strongly disagree, so a higher value means less agreement; perf3 is negatively worded (not confident of doing well) and is stored as answered, not reverse-scored, so it runs opposite to perf1 and perf2 (r = -0.51).
wallace_2026_s2_profcert	The items ask about ‘the previous questions’, which are the four ratings of the professor’s intelligence beliefs (IQB1, IQB2R, IQB3, IQB4) held in wallace_2026_mcm_check for Study 2.
wallace_2026_s2_verbalmath	Responses run 1 = Strongly agree to 7 = Strongly disagree, so a higher value means less agreement (e.g. a higher math1 means rating oneself as less good at math tasks).
wallace_2026_s2_warmcomp	Responses run 1 = Strongly agree to 7 = Strongly disagree, so a higher value means less agreement; warm3R (The professor seems mean.) is stored unreversed, so its high values mean the professor seemed less mean.
wallace_2026_s2c_covidcert	Responses run 1 = Strongly agree to 7 = Strongly disagree, so a higher value means less agreement that COVID-19 has created uncertainty.
wallace_2026_s2c_desirecert	Responses run 1 = Strongly agree to 7 = Strongly disagree, so a higher value means less desire for certainty; item desireforcert2 is the source file’s unnumbered column desireforcert.
wallace_2026_s2c_intoluncert	The five items are verbatim items 3, 5, 6, 15 and 25 of the Intolerance of Uncertainty Scale (IUS-27; Freeston et al., 1994; English version Buhr & Dugas, 2002). The wording ships on the distributor’s statement: PhenX (protocol 650701) describes the IUS as freely available with permission not required for use; the originators’ own terms were not located, and if they are found they govern. The study replaced the IUS anchors (1 = not at all characteristic of me to 5 = entirely characteristic of me) with its own reversed ones: responses run 1 = Describes me extremely well to 5 = Does not describe me, so a higher value means less intolerance of uncertainty, the reverse of the usual IUS scoring.
wallace_2026_s2c_mc	Responses run 1 = Strongly agree to 7 = Strongly disagree, so a higher value means less agreement; despite the MC item codes these are not manipulation-check items but four statements about reflecting on the certainty of one’s own attitudes.
wallace_2026_s2c_metacog	Responses run 1 = very much to 7 = not at all, so a higher value means less thinking about how the organization’s beliefs might influence one’s own.
wallace_2026_s2c_selfcert	Responses run 1 = very much to 7 = not at all, so a higher value means less certainty or confidence about oneself.
wallace_2026_s2c_uncertmind	Responses run 1 = Strongly agree to 7 = Strongly disagree, so a higher value means less agreement that one’s certainty or uncertainty is fixed.
wang_2015_donation_decision	Item wording is the paper’s English, but 208 of the 371 respondents were Chinese students who read an unpublished Chinese version (their donation question offered 100 Yuan rather than 100 dollars). Distress and sympathy ship with no option_text because the paper describes a 7-point scale from “not at all” to “very much” while the data store eight codes, 0-7, so the anchors cannot be placed.
wang_2016_study1_authalien	Item text is shown in English: Wang (2016) administered a Chinese translation of the Authenticity Scale and no Chinese wording is published in the PLOS deposit (the SPSS file carries no variable or value labels), so Wood et al.’s (2008) original English for Self- Alienation items 2, 7, 10 and 12 ships instead. IMPORTANT: these four items are stored ALREADY REVERSE-SCORED in the IRW table, so the response anchors shipped here run the opposite way to the study’s stated administration – resp = 1 is the respondent’s ‘strongly agree’ with the alienation statement and resp = 5 is ‘strongly disagree’. That direction is established from the data (all four items correlate +0.33 to +0.61 with the study’s self-esteem total, and the three authenticity tables sum to the paper’s published total of 42.49, SD 6.18, alpha .81), not assumed. Which of the four canonical items each code authalien1-4 refers to is inferred from the standard subscale numbering and could not be checked against the data – the block and its direction are verified, but a permutation within the four items would not be detectable.
wang_2016_study1_authexternal	Wang (2016) administered a Chinese translation of the Authenticity Scale and no Chinese wording is published in the PLOS deposit (the SPSS file carries no variable or value labels), so Wood et al.’s (2008) original English for Accepting External Influence items 3, 4, 5 and 6 ships instead. IMPORTANT: these four items are stored ALREADY REVERSE-SCORED in the IRW table, so the response anchors shipped here run the opposite way to the study’s stated administration – resp = 1 is the respondent’s ‘strongly agree’ with the external-influence statement and resp = 5 is ‘strongly disagree’. That direction is established from the data (all four items correlate +0.26 to +0.38 with the study’s self-esteem total, where the raw subscale must correlate negatively, and the three authenticity tables sum to the paper’s published total of 42.49, SD 6.18, alpha .81, against alpha .45 for the flipped reading), not assumed. Which of the four canonical items each code authexternal1-4 refers to is inferred from the standard subscale numbering and could not be checked against the data – the block and its direction are verified, but a permutation within the four items would not be detectable.
wang_2016_study1_authliving	Item text is shown in English: Wang (2016) administered a Chinese translation of the Authenticity Scale and no Chinese wording is published in the PLOS deposit (the SPSS file carries no variable or value labels), so Wood et al.’s (2008) original English for Authentic Living items 1, 8, 9 and 11 ships instead, with the study’s own 1 (strongly disagree) to 5 (strongly agree) anchors in place of Wood’s 1-7 anchors. Which of those four canonical items each code authliving1-4 refers to is inferred from the standard subscale numbering and could not be checked against the data – the block as a whole is verified (the three authenticity tables sum to the paper’s published total of 42.49, SD 6.18), but a permutation within the four items would not be detectable.
wang_2016_study1_power	Item text here is shown in English: Wang (2016) administered the authors’ own Chinese translation of the Sense of Power Scale and no Chinese wording is published anywhere in the PLOS deposit (the SPSS file carries no variable or value labels), so Anderson, John & Keltner’s (2012) original English ships instead. Two things to read carefully. The four reverse-worded items (power2, power4, power6, power7) are stored ALREADY REVERSE-SCORED – Cronbach’s alpha of the table as stored is .713 against the paper’s published .71, and un-reversing those four collapses it to .099 – so their option_text is deliberately flipped: resp 1 is ‘strongly agree’ with the printed statement and resp 5 ‘strongly disagree’, the opposite of the other four items. And which canonical item each code power1..power8 refers to is inferred from the standard Sense of Power numbering: the scale total (27.71, SD 3.78), its correlation with self-esteem (.61) and its alpha all reproduce the paper exactly, but a permutation within the eight items would not be detectable from the data.
wang_2016_study1_se	Item text for this table is shown in English: Wang (2016) administered a Chinese translation of the Rosenberg Self-Esteem Scale and no Chinese item wording is published in the PLOS deposit (none of the three SPSS files carries a variable label), so Rosenberg’s original English ships instead, in the item order of the edition the paper cites. The response options ARE the administered Chinese, taken from the same article’s Study 2 and Study 3 files, with an English translation written by IRW alongside; note that they are deliberately reversed for se2, se5, se6, se8 and se9, because the live responses are stored already reverse-coded (every pair of items correlates positively and the column sum reproduces the paper’s published total of 38.34, SD 5.08), so a 5 on those items means the respondent rejected the negative statement.
wang_2016_study3_auth	Item wording is shown in English: Wang (2016) Study 3 administered a Chinese version of the Authenticity Scale and no Chinese item wording is published anywhere in the PLOS deposit (the S3 SPSS file carries no variable labels), so Wood et al.’s (2008) original English ships in its place. The response options are different – those ARE the administered Chinese, taken verbatim from the same file’s value labels, with an IRW English rendering beside them. Which of the twelve canonical Authenticity Scale items each code a1-a12 refers to is inferred from the scale’s own numbering: the three subscale blocks are confirmed against the data (every item correlates more strongly with its canonical block than with either other block, and the scale total reproduces the paper’s published 38.38/6.88), but a permutation of the four items within a block would not be detectable
wang_2016_study3_power	Item wording here is shown in English: Wang (2016) Study 3 administered a Chinese translation of the Sense of Power Scale and no Chinese item wording is published anywhere in the PLOS deposit (the S3 SPSS file carries no variable labels), so Anderson, John & Keltner’s (2012) original English ships in its place. The response options are different – those ARE the administered Chinese, taken verbatim from the same file’s value labels, with an IRW English rendering beside them. Read two things carefully. The four reverse-worded items (p2, p4, p6, p7) are stored ALREADY REVERSE-SCORED – the raw column sum reproduces the paper’s published power total of 29.37 (SD 5.37) exactly and alpha is .878 against the published .88, while un- reversing those four drops alpha to .441 – so their option text is deliberately flipped: a 5 means the respondent rejected the printed negative statement. And which canonical Sense of Power item each code p1-p8 refers to is inferred from the scale’s standard numbering; the forward/reverse split is confirmed against the data (rank 1 of all 35 possible 4/4 splits on the wording method factor), but a permutation of items within either the forward or the reverse group would not be detectable.
wang_2016_study3_se	Item wording for this table is shown in English: Wang (2016) Study 3 administered a Chinese translation of the Rosenberg Self-Esteem Scale and no Chinese item wording is published anywhere in the PLOS deposit (the S3 SPSS file carries no variable labels), so Rosenberg’s original English ships instead, in the item order of the edition the paper cites. The response options ARE the administered Chinese, taken verbatim from that file’s value labels, with an English rendering written by IRW alongside. Note that the anchors are deliberately reversed for se2, se5, se6, se8 and se9: the responses are stored already reverse-coded (all 45 inter-item correlations are positive and the column sum reproduces the paper’s published 39.12, SD 6.95, with alpha .898 against a published .90), so a 5 on those items means the respondent rejected the negative statement. Which of the ten canonical RSES sentences each code se1-se10 names is inferred from that edition’s numbering; the split into the five negatively and five positively worded items is confirmed against the data (rank 1 of all 126 possible 5/5 splits), but a permutation within either group would not be detectable.
wang_2025_green_space_wellbeing	The English item text for this table was translated from the administered Chinese by IRW, not by the study authors; the Chinese wording respondents actually read ships alongside it in item_text. The study published no response-option labels for its 1-5 scale, so option_text is blank
wang_2026_attitude	Item wording for this table is shown in English. Wang & Zou administered a Chinese translation and back-translation of the attitude scale and no Chinese wording is published anywhere in the deposit, so the English from the study’s own S1 Appendix questionnaire ships in its place. Which of the three items each code ATT1-ATT3 refers to follows the appendix’s own AT1/AT2/AT3 numbering; the data confirm the paper’s codes sit on these columns (recomputed loadings 0.824/0.818/0.834 against the published 0.825/0.815/0.836, with alpha, CR and AVE exact), but the three items are near-synonymous enjoyment statements with means of 3.79, 3.72 and 3.73, so a permutation among them would not be detectable from the responses.
wang_2026_behavioral_intention	Item and response wording for this table is shown in English. Wang and Zou (2026) administered a Chinese translation of these behavioural-intention items and no Chinese wording is published anywhere in the study’s deposit – the S1 Appendix questionnaire and the S2 Appendix data file both contain zero Chinese characters – so the authors’ own English from the S1 Appendix ships in its place. Which item each code BI1-BI5 refers to is not in doubt: the appendix prints the code as a prefix to every item, and each live item’s response-level counts reproduce the identically- named deposit column exactly.
wang_2026_perceived_usefulness	Item and response wording for this table is shown in English. Wang and Zou (2026) administered a Chinese translation of these perceived-usefulness items and no Chinese wording is published anywhere in the study’s deposit – the S1 Appendix questionnaire and the S2 Appendix data file both contain no Chinese characters – so the authors’ own English from the S1 Appendix ships in its place. Which item each code PU1-PU3 refers to is not in doubt: the appendix prints the code as a prefix to every item, each live item’s response-level counts reproduce the identically-named deposit column exactly, and the paper’s published per-item VIFs (1.626 / 1.554 / 1.500) recompute from those columns to three decimal places.
wang_2026_technology_perception	This table was published until 2026-09-27 as wang_2026_teaching_presence. It was renamed because its items are the study’s Technology Perception (TP) construct, teachers’ judgements of digital games in the classroom; “teaching presence” appears nowhere in the article or its appendix. The wording is English: Wang and Zou (2026) administered a Chinese translation, and no Chinese wording appears anywhere in their deposit. The appendix numbers twelve TP items, while the data carry eight, renumbered after four low-loading items were dropped. The shipped text follows the assignment that reproduces all eight of the paper’s published Table 2 VIFs to within 0.0005, the unique best of 40,320 possible assignments. Two items are negatively worded and stored already reverse-scored: TP3 (‘Requires DG technological training, which takes too much time.’) and TP5 (‘Makes classroom management more difficult.’). For those two only, resp 1 means Strongly Agree and resp 5 means Strongly Disagree
wang_2026_veteran_expectations	Item and response wording for this table is shown in English. Wang and Zhang (2026) surveyed veterans in Guizhou Province, China, and never state the administered language, but the questionnaire cannot have been given in English; no Chinese wording appears anywhere in the article or its single deposited file, so the authors’ own English questionnaire (the paper’s Table 1) ships in its place. That table is published as an image, so the item text is a transcription read from the image rather than copied as text. Which item each code B1-B3 refers to is not in doubt: Table 1 prints the data’s own code as a label on every item, and each live item’s counts of responses 1-7 reproduce the identically-named deposit column exactly (B1 0,6,53,120,169,165,111; B2 10,31,105,166,148,99,65; B3 9,35,95,170,168,107,40). One caveat about the response data rather than the wording: the paper analyses 525 respondents while the deposited file and this table hold 624, and the published Cronbach’s alpha of 0.76 for these three items recomputes to 0.66 on the 624.
wang2022_dass_anxiety	Item text for this table is shown in English: Wang et al. (2022) administered the Chinese DASS-21 (Wang K et al., 2016) and no Chinese wording is published in the study’s deposit – the figshare .xlsx labels its DASS columns DASS1..DASS21 with no variable or value labels – so Lovibond & Lovibond’s original English for the seven anxiety items ships instead. The response options are the study’s own English for its endpoints only (‘0 = completely inconsistent’, ‘3 = completely in line’); 1 and 2 are blank because the paper labels no middle anchors, and the canonical DASS anchors were deliberately not substituted because this administration relabelled the scale. Which canonical anxiety item each code dass_anx_1..7 refers to is inferred from the deposit’s column numbering: the subscale itself is confirmed against the data (alpha 0.8436 for these seven, 0.8729 and 0.8602 for the depression and stress triples, against the paper’s published 0.84 / 0.87 / 0.86) and each item reproduces one deposit column exactly, but a permutation of the seven sentences among those columns would not be detectable.
wang2022_dass_depression	Item text for this table is shown in English: Wang et al. (2022) administered the Chinese DASS-21 (Wang K et al., 2016) and no Chinese wording is published in the study’s deposit – the figshare .xlsx labels its DASS columns DASS1..DASS21 with no variable or value labels – so Lovibond & Lovibond’s original English for the seven depression items ships instead. The response options are the study’s own English for its endpoints only (‘0 = completely inconsistent’, ‘3 = completely in line’); 1 and 2 are blank because the paper labels no middle anchors, and the canonical DASS anchors were deliberately not substituted because this administration relabelled the scale. Which canonical depression item each code dass_dep_1..7 refers to is inferred from the deposit’s column numbering: the subscale itself is confirmed against the data (alpha 0.8729 for these seven against the paper’s published 0.87, and all twelve per-gender cells of the paper’s Table 1 reproduce to two decimals) and each item’s full response distribution matches exactly one deposit column and no other, but a permutation of the seven sentences among those columns would not be detectable.
wang2022_dass_stress	Item text for this table is shown in English: Wang et al. (2022) administered the Chinese DASS-21 (Wang K et al., 2016) and no Chinese wording is published in the study’s deposit – the figshare .xlsx labels its DASS columns DASS1..DASS21 with no variable or value labels – so Lovibond & Lovibond’s original English for the seven stress items ships instead. The response options are the study’s own English for its endpoints only (‘0 = completely inconsistent’, ‘3 = completely in line’); 1 and 2 are blank because the paper labels no middle anchors, and the canonical DASS anchors were deliberately not substituted because this administration relabelled the scale. Which canonical stress item each code dass_str_1..7 refers to is inferred from the deposit’s column numbering: the subscale itself is confirmed against the data (alpha 0.8602 for these seven against the paper’s published 0.86, and the paper’s Table 1 stress means by gender and grade reproduce to two decimals) and each item reproduces one deposit column exactly, but a permutation of the seven sentences among those columns would not be detectable
warwas_2022_sharing_economy	The survey was conducted by telephone in Polish with a representative sample of adult Polish residents, but the only questionnaire published with the study is written in English, so the item text shipped here is that English rendering rather than the Polish respondents heard
weatherspoon_2015_family_physicians_effectiveness	The survey’s effectiveness column offered only Yes (1), No (2) and Don’t know (9, dropped in processing), but the live table contains 25 responses coded 3, 4 or 5; those rows have no option text. Item wording was transcribed from an image of the questionnaire’s Q12 grid.
weber2026_name_knowledge	item_text, option_text and instructions carry the administered German; the English in the _translated columns is the authors’ own, published in the same record, rather than an IRW rendering. The 1-5 response direction follows the order the printed questionnaire shows its checkboxes in: the study’s data file stores unlabelled integers with no value labels, so the direction could not be confirmed against the data and rests on the instrument alone.
weeldenburg_2022_targetmtpq	Item wording for this table is shown in English. Weeldenburg et al. (2022) administered the Mastery Teaching Perception Questionnaire in Dutch to 3,150 Dutch secondary-school students, and no Dutch wording is published anywhere in the deposit – the SPSS file labels its 26 item columns with bare internal codes (TA_1, AU_1, RE_1, …) and the only questionnaire supplement is the authors’ own English version, which is what ships. Response options carry only the two endpoints (‘strongly disagree’, ‘strongly agree’); 2, 3 and 4 are blank because the source labels no middle anchors. Which of the 26 numbered questionnaire items each code refers to is inferred within each TARGET construct from the questionnaire’s block- interleaved numbering: construct membership is confirmed against the data (each block reproduces the paper’s Table 1 exactly – Task 3.56/.70/alpha .78, Authority 3.00/.85/.79, Recognition 3.66/.79/.85, Grouping 3.37/.81/.80, Evaluation 3.44/.79/.82) and two Authority positions are confirmed by content (Authority_2 is the only Authority item correlating most with the Evaluation block, Authority_5 the only one correlating most with Grouping), but a permutation of items inside a construct block would not be detectable.
weida_2020_cesd10	This table was published until 2026-09-27 as weida_2020_financial_security. It was renamed because its ten items are the study’s CES-D-10 depression scale: summing them as stored reproduces the authors’ own depression score exactly (mean 11.40, SD 6.20, range 0-30, 56.3% at or above the published cutoff of 10). The item wording is the canonical CES-D, not wording documented from this study’s own materials; the deposit’s columns carry no labels. The CES-D is shipped on second-hand statements that it is free to use (“This scale is free to use without permission”, as quoted by CamCOPS). No restriction from its originator (Radloff, NIMH) was found. The two positive-affect items, secf_5m and secf_8m, are stored already reverse-scored, so their anchors run opposite to the other eight. secf_4m is off-scale in this sample (item-rest correlation 0.07, against 0.27-0.73 for the rest), and it is the least certain of the ten code-to-item assignments
wen_2022_pyd	Item wording is the authors’ own published English; respondents read a Chinese questionnaire whose wording the study never deposited, and the 1-5 response anchors are not published anywhere, so option_text is blank.
wesselmann_2018_distress	The paper prints only the two endpoint anchors of the 1-6 scale (‘All of the time’, ‘Never’) and the deposited SPSS file carries no value labels at all, so option_text is blank for the four middle points rather than filled in from the canonical MHI wording this study renames; the instructions field carries the instrument’s standard item stem (‘How much of the time during the past month have you…’), which the paper corroborates but never prints
wesselmann_2018_drri	The five response labels come from the published DRRI-2 form rather than from the study, because the deposited SPSS file carries no value labels and the authors’ own codebook misprints the anchor for 1 as ‘Strongly Agree’ (it is ‘Strongly disagree’, as the paper states). Item wording is the deposited file’s own variable labels, shipped as administered; it differs from the DRRI-2 form only in omitting the form’s shared ‘During deployment…’ stem and the final period on drri2.
wesselmann_2018_pcl	The item text is the deposited file’s own variable labels, which differ from the current PCL-5 form at three items (pcl9 ‘no one else can be trusted’, pcl14 ‘having loving feelings’, pcl17’s quotation marks) and are shipped as administered rather than normalised; the instructions and the five response labels come from the published PCL-5 form, because the study’s own codebook prints another scale’s prompt and response options against its PCL-5 block
wesselmann_2018_social_support	The response anchor for 1 is shipped as ‘Strongly Disagree’ although the authors’ own codebook prints it as ‘Strongly Agree’ – a misprint that duplicates the anchor for 5, corrected against the paper’s ‘1-Strongly Disagree, 5-Strongly Agree’ and the published DRRI scale page; item wording is the deposited SPSS file’s own variable labels, shipped as administered
wiedemann_2021_risk_perception	The item stems are the administered German from the study’s SPSS labels, but the scale anchors and instructions are the study’s English translation because the German originals are not in the deposit. The items rate a hazard-communication text that differed by experimental condition (see cov_risk_type, cov_factor1, cov_factor2), so no single passage is shipped.
williams_2022_dvmsq	The wording is the REDCap form this study administered, which is wider than the published 18-item DVMSQ v1.5: misophonia_overreact and miso_imp_conc are not on the printed form, miso_fear is printed (item 3) but not counted in the published scores, and the three global-impact items’ second option reads ‘A little’ where the printed form has ‘A little bit’. Respondents who answered No to misophonia_screen (570 of 1,403) were not shown the other items and have only that row.
wine_luckett2021	Item text is the raw data file’s column header, not the question raters were asked: “Pleasantness” and “Familiarity” are one-word labels, and only intensity is a full question. The three items use different scales – intensity a 0-100 line scale, pleasantness and familiarity 9-point scales – and no anchors ship. id is the wine sample, not a person; the judge is in rater.
wirs	The item wording is transcribed from the documentation of the R package that distributes this dataset rather than from a questionnaire form, and the source never states the direction of the 0/1 coding or publishes response labels, so option_text is blank throughout
WMD_bartczuk_2023	The Polish form’s instructions say it contains 28 statements although it prints (and the data hold) 24 items, and the English for the instructions and response anchors is IRW’s own translation; item English is the study’s.
wolf_2017_study1_posaff	Despite the table name and dictionary description, these 12 items are a 12-adjective selection from the full PANAS and include six negative-affect adjectives (distressed, upset, guilty, scared, hostile, irritable) as well as six positive ones; the stored responses are raw, so the negative items must be reverse-scored before a positive-affect score is computed.
wolf_2017_study2_posaff_post	Despite the table name and description, these 12 items are a 12-adjective selection from the full PANAS and include six negative-affect adjectives (distressed, upset, guilty, scared, hostile, irritable) as well as six positive ones; the stored responses are raw, so the negative items must be reverse-scored before a positive- affect score is computed.
wolf_2017_study2_posaff_pre	Despite the table name and description, these 8 items are an 8-adjective selection from the PANAS and include four negative-affect adjectives (upset, scared, hostile, irritable) as well as four positive ones (proud, enthusiastic, interested, alert); the stored responses are raw, so the negative items must be reverse-scored before a positive-affect score is computed.
wolf_2017_study2_video_liking	Despite the table name, this 6-item scale is not described, named or reported anywhere in the published paper – it exists only in the study’s Study 2 data deposit, whose SPSS variable labels are the sole source of the wording, and it mixes liking (‘I liked the video’) with calm, happiness and satisfaction; three items (like2, like4, like6) are negatively worded and are stored RAW, so they must be reverse-scored before a liking score is computed (doing so reproduces the deposit’s own video_like composite exactly).
wolters2026_panas_na	Respondents completed the German PANAS (Krohne et al., 1996); the wording IRW ships is the English original (Watson, Clark & Tellegen, 1988) mapped item-for-item onto it, because the deposit contains no item wording at all and the German version is published only under non-commercial, permission-required terms that IRW does not redistribute under
wolters2026_panas_pa	Respondents completed the German PANAS (Krohne et al., 1996); the wording IRW ships is the English original (Watson, Clark & Tellegen, 1988) mapped item-for-item onto it, because the deposit contains no item wording at all and the German version is published only under non-commercial, permission-required terms that IRW does not redistribute under
wolters2026_wi	Item wording for this table is shown in English: the study was run in German and no German wording is published in its figshare deposit (a plain spreadsheet whose only WI content is the column names WI_1..WI_14), so the Whiteley Index’s English, as reproduced by Eslami, Ahmadipour & Bagheri (2017), ships instead. Two things to read carefully. Which of the fourteen canonical Whiteley items each code wi_1..wi_14 refers to is inferred from the instrument’s own numbering, not stated by the source: the item-rest profile matches published per-item loadings at Spearman 0.54 (permutation p = 0.026), which rules out a scrambled mapping but would not detect a swap between two items of similar loading. And the responses are Whiteley SCORES, not raw answers, so for wi_9 – the scale’s one reverse-keyed item – a 1 means the respondent answered ‘no’, the opposite of every other item; its item-rest correlation is +0.32 where raw storage would be negative.
wooly_hartman2022	The responses here are right/wrong scores rather than the alternative each person chose. On the four TV-show items (Q4, Q6, Q8, Q75) the source gives ‘I Don’t Know’ no code, so those responses are absent rather than scored 0 – only 1023-1286 of the 1561 respondents carry a score on those items – whereas on the 15 song items ‘I Don’t Know’ is scored incorrect
wu_2015_bsrs5	The questionnaire was administered in Chinese (Taiwan), but the source publishes its item wording only in English: the item text is the authors’ English rendering, taken from the variable and value labels of their deposited data file, and the Chinese wording respondents read is not available. Item BSRS6 is the suicide-ideation item the study added to the five-item BSRS-5.
wu_2015_cias	The questionnaire was administered in Chinese (Taiwan), but the source publishes its item wording only in English: the item text is the authors’ English rendering, taken from the variable and value labels of their deposited data file, and the Chinese wording respondents read is not available.
wu_2024_resilience	wu_2024_resilience was administered in Chinese (the Chen et al. 2020 Chinese Brief Resilience Scale), but the Chinese wording is published nowhere in the study’s deposit or supplements, so the item and option text shipped here is Smith et al.’s (2008) original English BRS rather than what respondents read; the instruction line is likewise the BRS’s printed paper-form instruction, while this study was administered online
wu_2024_video_addiction	Item codes D1-D14 carry no wording in the source file, so their match to Qin et al.’s (2019) items was reconstructed from the data: the four subscales and the seven diagnostic items are pinned by reproducing the study’s published alphas and its count of 132 addicts, but order within D1/D3, D2/D4/D5, D6/D8 and D9/D10/D11 is not established; the Chinese wording is Qin et al.’s Table 1 rather than a text the study published, and English for seven items and all response anchors was machine-translated by IRW.
wu_2025_drone_delivery	The survey was run in Kunming, China, on a Chinese survey platform, so it was almost certainly administered in Chinese (the study does not say); no Chinese wording is published, and the item text is the authors’ own English from the deposit. The five Service Performance items (Q30-Q34) are bare attribute phrases (‘Distribution costs’, ‘Delivery speed’, …) because the stem respondents rated them against was never published. Only the endpoints of the 5-point scale are labelled.
wu2021_basic_empathy	Respondents answered the 16-item Chinese Basic Empathy Scale; item_text is that Chinese as administered. The English is Jolliffe and Farrington’s original wording for the matching BES-20 item (the Chinese form drops items 2, 3, 4 and 15), not a translation of the Chinese. The eight negatively worded items (D1, D3, D4, D5, D10, D14, D15, D16) are stored already reverse-recoded, so for them resp 1 = ‘completely agree’ and 5 = ‘completely disagree’.
wu2026_liking	The ten items of this table are ten ceramic vase photographs, not ten different questions: every item carries the same single statement, ‘This vase is pleasing to see.’, and the vases themselves (Fig 1 of the source paper) are images with no published descriptions, so nothing in the item text distinguishes one item from another. The English shown is the paper’s own wording; the survey was run face-to- face with 200 native Chinese speakers and no Chinese version of the statement appears anywhere in the deposit.
wu2026_novel	Each of the ten items rates a different photographed ceramic vase, so all ten share the same single-item wording “This is a novel vase.” and the stimulus is identified only by the item code: Novel_k is vase k of Fig 1 of the source paper. The shipped wording is the paper’s English; the survey was run face-to-face with native Chinese speakers and no Chinese wording is published in the deposit or supplements. Only the 1 and 7 scale points are labelled, as in the source.
wu2026_typical	The ten items of this table are ten ceramic vase photographs, not ten different questions: every item carries the same single statement, ‘This is a typical vase.’, and the vases themselves (Fig 1 of the source paper) are images with no published descriptions, so nothing in the item text distinguishes one item from another. The English shown is the paper’s own wording; the survey was run face-to-face with 200 native Chinese speakers and no Chinese version of the statement appears anywhere in the deposit.
wurm_2016_mdi	The Wurm et al. physicians were surveyed in German, but the German MDI wording is in neither the deposited SPSS file nor the paper’s supplements, so the item text shipped here is the canonical English MDI (Nielsen, Oernboel, Bech & Christensen 2019, Table 2) rather than what respondents read. Items mdi8_neu and mdi10_neu are each scored as the higher of two sub-items – restless vs slowed down, and reduced vs increased appetite – so their item_text carries both alternatives and a response of 4 does not say which of the two produced it.
xiao_2024_hierarchical_plateau	Item codes HP1-HP4 are assumed to follow the questionnaire’s item order (the source data file has no item labels and the paper publishes no per-item statistics to confirm it); the English response labels are the study’s own rendering, while the Chinese anchors administered read ‘very untrue/very true of me’ rather than disagree/agree.
xiao_2024_taking_charge	Item codes TC1-TC4 were matched to the questionnaire’s four items by their printed order, since the data file carries no labels and no per-item statistics are published to confirm it; the article describes agree/disagree anchors while the questionnaire respondents answered prints frequency anchors, which are what is shipped.
xiao_2026_entrepreneurial_intention	The shipped item text is the authors’ own English; respondents answered a Chinese version whose wording the study does not publish. The four entrepreneurial-intention items are In1-In4 in the data but EI1-EI4 in the study’s codebook, and their one-to- one correspondence follows the codebook’s listing order rather than any stated key.
xiong_2025_dass21	The DASS-21 instructions and 0-3 anchors shipped for this table are the official paper form’s wording (“Please read each statement and circle a number 0, 1, 2 or 3”); Xiong et al. administered the scale online through Google Forms, so respondents selected rather than circled a number. The item wording is the same official form’s, verbatim.
xu_2022_ples_aa	The wording shipped for item_1.1 is the deposited questionnaire’s own header, ‘Field of Higher Education, Implement PLE Platform’s information disclosure policy.’, which differs in substance from item 1.1 of the final published PLES-AA (‘Provide a rationale for the implementation of PLE in Higher Education’, S1 Appendix); the other 15 items agree between the two. Separately, the survey was administered in Chinese via WenJuanXing and no Chinese wording appears anywhere in the deposit or supplements, so all item text here is the study’s own English.
xu_2025_hcq_p	For the 23 items Xu et al. list as reverse-scored (HCQ-P 2, 5, 6, 12, 13, 14, 17, 19, 21, 22, 24, 25, 26, 27, 30, 32, 34, 38, 39, 40, 43, 45, 48) the stored resp is already reverse-scored, so their shipped anchors run 1 = Strongly agree to 6 = Strongly disagree while the remaining 26 items run the other way; separately, the item wording is the administered Chinese but the two anchor labels are the paper’s English, because no Chinese anchor wording is published and scale points 2-5 are unlabelled.
xu_2026_game_experience	The English item text (item_text_translated) was written by IRW from the Chinese column headers of the deposit, which carry the stems respondents read; it is not the authors’ wording.
xu_2026_participation	As for xu_2026_game_experience: IRW’s English translation of the Chinese headers.
xu_2026_substitutability	As for xu_2026_game_experience: IRW’s English translation of the Chinese headers.
yandun2026_attention	The item wording shipped for this table is the authors’ own English translation, deposited as DATA_MATRIX_TRANSLATED.xlsx; the assessment itself was administered in Spanish in Ecuador and no Spanish wording was deposited. The 1-5 anchors are published only at the endpoints (1 = Never, 5 = Always), so options 2-4 carry no text.
yandun2026_logical_thinking	The item wording shipped for this table is the authors’ own English translation, deposited as DATA_MATRIX_TRANSLATED.xlsx; the assessment itself was administered in Spanish in Ecuador and no Spanish wording was deposited. The 1-5 anchors are published only at the endpoints (1 = Never, 5 = Always), so options 2-4 carry no text.
yandun2026_memory	The item wording shipped for this table is the authors’ own English translation, deposited as DATA_MATRIX_TRANSLATED.xlsx; the assessment itself was administered in Spanish in Ecuador and no Spanish wording was deposited. The 1-5 anchors are published only at the endpoints (1 = Never, 5 = Always), so options 2-4 carry no text.
yang_2018_cesd	The Yang et al. respondents answered a Chinese CES-D, but no Chinese wording appears in the deposited SPSS file or the paper’s supplements, so the item text shipped here is the canonical English Radloff (1977) CES-D and the response anchors are the authors’ own English labels from that file; the items are stored raw (resp 1 = least frequent) for all twenty, including the four positively worded ones (E4, E8, E12, E16), which the authors reverse before summing.
yang_2023_consumption_intent	The wording shown is English: the survey was administered in Chinese and no Chinese wording appears in the article’s single supplementary file, so the authors’ own English from Table 1 is what ships, transcribed from a table published only as an image; item E3 ends mid-phrase (‘…information and knowledge of’) in the source and was not repaired.
yang_2023_emotional_eating_cesd	The respondents were Chinese undergraduates answering a Chinese CES-D, but the study’s single supplementary file contains no Chinese text at all, so the item wording shipped here is the authors’ own English column headers from that file (rough, unpunctuated, and rendering both CES-D items 6 and 18 as ‘I feel sad’). One item, CESD_2, has no header wording of its own – the survey export overwrote it with the block instruction – so it carries the canonical English CES-D item 1 wording, identified by elimination and corroborated against the file’s own precomputed total score. Note also that CESD_1 and CESD_2 are canonical CES-D items 2 and 1 in that order, not 1 and 2. No response-option wording is shipped: the source publishes none and the paper’s description of the response scale contradicts itself.
yang_2023_emotional_eating_ders	The survey was administered in Chinese and no Chinese wording is published in the article’s single supplementary file, so the item text shipped here is the English column headers of the authors’ own deposited workbook, transcribed verbatim including their rough phrasing and typos. Two rows are not the study’s words: item DERS_1, whose source column header was the block’s instruction text rather than an item, carries the canonical Gratz & Roemer (2004) DERS item 1 placed by elimination; and the five response anchors, which are the canonical DERS labels (the paper misstates the format as 1-4 while the data carry five levels). Responses are stored RAW, so the 11 reverse-keyed items (DERS_1, 2, 6, 7, 8, 10, 17, 20, 22, 24, 34) must be reversed before summing a DERS total.
yang_2023_emotional_eating_eesr	The wording shown for this table is English taken from the column headers of the study’s own deposited data file; the scale was administered in Chinese (Zhu Hong’s 2012 Chinese revision of the Emotional Eating Scale) and no Chinese wording is published in the article’s single supplementary file. Two consequences a user should know: item EESR_1 carries no text at all, because the survey export overwrote that column’s label with the question stem shown in instructions, and two pairs of distinct items share an identical English label (EESR_5 and EESR_20 both read ‘hostile’; EESR_13 and EESR_22 both read ‘depressed’) because the deposit’s English does not distinguish the underlying Chinese words. The 1-5 response options carry no text: the paper describes only the endpoints (1 = a total lack of the emotion, 5 = a high degree) and never publishes anchor labels.
yang_2023_emotional_eating_uppsp	The scale was administered in Chinese and the article’s single supplementary file contains no Chinese text, so the wording shown here is the English column headers of the study’s own deposited workbook, transcribed verbatim including their rough phrasing and their mid-sentence truncations (UPPSP_12 and UPPSP_14 simply stop part- way). Two things are not the study’s words: UPPSP_1, whose header the survey export overwrote with the block instruction and truncated to the single letter ‘I’, carries the canonical SUPPS-P item ‘I generally like to see things through to the end.’ placed by elimination on the scoring key and corroborated in the data; and the two response anchors, which are shipped 1 = strongly disagree, 4 = strongly agree – the REVERSE of the paper’s Measures sentence, because the negative-urgency items correlate +0.54 with the study’s own depression total, which is only possible if the high codes are the agreeing pole. The two middle levels carry no text because the study publishes none. Responses are stored raw, so the premeditation and perseverance items (UPPSP_1, 2, 4, 5, 7, 11, 12, 19) must be reversed before summing an impulsivity total.
yang_2023_green_brand_image	The wording shown is English: the survey was administered in Chinese and no Chinese wording is published in the article’s single supplementary file, so the authors’ own English from the paper’s Table 1 is what ships. One item code needs care: the paper’s tables print the third social-image item as C3, while the study’s raw data file (and therefore this table) calls that column C4 – the item shipped for C4 is the one the paper labels C3, confirmed by reproducing the paper’s published reliability for that three-item factor (.712) from C1, C2 and C4 exactly. Only the two endpoints of the 1-5 scale are published (1 = disagree, 5 = agree), so options 2-4 carry no text.
yang_2023_perceived_value	The item wording shown is English: the survey was administered in Chinese and no Chinese wording is published in the article’s single supplementary file, so the authors’ own English from Table 1 is what ships, transcribed from a table published only as an image.
yang_2025_intercultural_contact	The item wording, instructions and response anchors are transcribed verbatim from the study’s own questionnaire (S1 File of the CC BY article), including its typos – item FICA5 reads ‘lving in a homestay familiy abroad’ as printed. Two things are worth knowing. First, the anchors shipped here are Never / Rarely / Sometimes / Often / Always, which is what the questionnaire prints above every item of this scale and what its instruction to answer ‘based on the frequencies of your engaging in’ implies; the paper’s Methods section instead describes Sections II and III as running ‘1 (strongly disagree) to 5 (strongly agree)’, which matches neither section’s printed anchors, so a reader comparing the two will find a discrepancy the questionnaire settles. Second, the workbook’s column headers are bare codes (FSM1, FICA1, …) and the questionnaire numbers its lines under the block headings ‘Foreign Social Media (FSM)’ and ‘Foreign Intercultural Communication Activity (FICA)’, so the block each item belongs to is certain but the order of the three FSM texts among themselves, and of the six FICA texts among themselves, follows that printed numbering and could not be checked against the data: the paper reports no per-item statistics and the nine item means span only 2.88 to 3.22. What was checked: all 45 item-by-response-level counts in the deposited workbook match the live table exactly, so no code was permuted in processing and the 1-5 integers are the source’s own.
yang_2026_igd_benefits	Despite the table name, only two of the four items rate perceived BENEFITS of internet gaming (benefit_a = at present, benefit_c = in the future); benefit_b and benefit_d rate perceived COSTS at present and in the future, which the shipped item text states. Only the scale endpoints were published (1 = none, 10 = extremely strong), so response options 2-9 carry no text.
yang_2026_pars3	Item wording for yang_2026_pars3 is the source paper’s own English rendering of a Chinese-administered PARS-3, and the paper introduces each stem with ‘e.g.’, so the stems are illustrative rather than a verbatim transcript; the administered Chinese wording is not published anywhere in the deposit.
yang2026_efl_learner_perspective	The four rating items (item_29-item_32) are stored with 1 = strongest agreement and 4 = weakest, the reverse of the 1=lowest/4=highest scoring the source paper reports; the paper’s means 2.98/2.96/3.08/3.21 equal 5 minus the stored means. No response- option wording is shipped because neither the deposit nor the paper publishes anchors, and the 0/1 items are select-all-that-apply checkboxes where 1 = option selected.
yao_2020_gad	The GAD-7 was administered in Chinese, but no Chinese wording is published in the study’s deposit or reachable in its paper, so the item and option text is the publisher’s official English GAD-7. The data carry no item labels, so which code is which item rests on the standard GAD-7 numbering: items 4 and 6 are confirmed from the data, but items 1-3, and items 5 and 7, cannot be told apart from it.
yao_2020_ius	The IUS was administered in a Chinese translation, but no Chinese wording is published in the study’s Mendeley deposit or reachable in its paper, so the item and option text shipped here is the English version (Buhr & Dugas, 2002) as distributed by PhenX rather than the sentences respondents read.
yao_2020_pswq	The PSWQ was administered in Chinese (Zhong et al. 2009 version), but no Chinese wording is published in the study deposit or reachable paper, so the text shipped here is the English original as reproduced by CamCOPS; items pswq1, pswq3, pswq8, pswq10 and pswq11 are reverse-worded and are stored unreversed in the response table.
yin_2022_gad7	The seven item codes in this table are the source workbook’s own column headers (‘GAD-7 question: afraid’ and so on), which name each symptom but are not the sentences respondents answered; the item text, instructions and the four response labels shipped here are the official GAD-7 English form’s verbatim wording. The study’s own Methods section independently reproduces all seven stems word for word in the same order as the columns, so only capitalisation and the form’s ‘(Use a tick to indicate your answer)’ line differ – that line is dropped because this study used a Google Form. The paper prints only the endpoints of the 0-3 scale, so the middle anchors ‘Several days’ and ‘More than half the days’ also come from the official form. Separately, ‘GAD-7 question: trouble relaxing’ has 151 responses where the other six have 153, from two blank cells in the source file.
yoshimura_2026_es_scale	The study did not release its questionnaire, so the item wording is the published 1994 Japanese scale (Komaki) that the study cites, which the study may have adapted to refer to the supervisor. The English for item 1 is the study’s own; the English for items 2-8 was translated by IRW. The response labels are the study’s English codebook anchors, not the Japanese wording respondents saw.
yoshimura_2026_isu_scale	The study did not release its questionnaire, so the item wording is the published 1994 Japanese scale (Komaki) that the study cites, which the study may have adapted to refer to the supervisor. The English for item 9 is the study’s own; the English for items 10-14 was translated by IRW. The response labels are the study’s English codebook anchors, not the Japanese wording respondents saw.
yoshimura_2026_psych_safety	The study did not release its questionnaire, so the item wording is the published Japanese version of the scale (Sasaki et al. 2022) that the study cites, and the instructions are that version’s. The English alongside is O’Donovan et al.’s original, a parallel version rather than a translation, so it does not always match the Japanese word for word (item 7: ‘my input is valued’ vs ‘listens to it with respect’). Item order within three groups of three items is inferred from numbering and not independently verified.
yu_2025_bedtime_procrastination	Item text for yu_2025_bedtime_procrastination is the canonical English Bedtime Procrastination Scale (Kroese et al. 2014); the study administered a Chinese revision whose wording is published nowhere in the deposit or the article. Two caveats on the response options: the source paper’s own anchor list prints ‘sometimes’ twice (1 = never; 2 = sometimes; 3 = sometimes; 4 = often; 5 = always) and is transcribed as printed, and for BP2, BP3, BP7 and BP9 the stored resp is the already-reverse-scored value rather than the administered answer, so those four items carry the anchors in descending order (resp 1 = always … 5 = never).
yu_2025_physical_activity	Item text for yu_2025_physical_activity is a third-party paper’s published English rendering of the Chinese-administered PARS-3 (Zhang & Zhang 2022, IJERPH 19:11451, Table 1), not the wording the study’s adolescents read, and the assignment of PE1/PE2/PE3 to intensity/duration/frequency is inferred from the paper’s stated dimension order rather than verified against the data – the source publishes no per-item statistics, and every PA-block statistic it does publish (alpha 0.828/0.688/0.730, KMO) is unchanged by any relabelling of the three items.
yuan_2024_contact	As for yuan_2024_ucla: the deposit’s English wording, not the Chinese wording respondents read.
yuan_2024_trust	As for yuan_2024_ucla: the deposit’s English wording, not the Chinese wording respondents read.
yuan_2024_ucla	The item text comes from the English SPSS labels in the deposit, the authors’ rendering of a questionnaire administered in Chinese to secondary-school students in Sichuan; the Chinese wording respondents read is not published.
yuebo_2024_netbenefit	Item text for yuebo_2024_netbenefit is the study’s own English version of a questionnaire that was administered in Chinese; the administered Chinese wording is published in neither the paper nor its supplements. No response-option labels are shipped because the source states only that the items used a 5-point Likert scale and prints no anchors.
zamzuri_2021_rpap_attitude	Respondents completed a bilingual English/Malay form; only the English wording is published, so the Malay text they may have read is not included.
zamzuri_2021_rpap_practice	Respondents completed a bilingual English/Malay form; only the English wording is published, so the Malay text they may have read is not included.
zamzuri_2021_rpap_risk_perception	Respondents completed a bilingual English/Malay form; only the English wording is published, so the Malay text they may have read is not included. In the response data, items D2, D3, D7, D8, D9, D10, D11 and D12 are stored reverse-scored (1 = strongly agree), so their anchors here are reversed to match; D1, D4, D5 and D6 keep 1 = strongly disagree.
zautra_2015_tmms24	Item wording is transcribed verbatim from the study’s own codebook, including its spelling slips (e.g. ‘attention’ and ‘definer’ in items 8 and 10 for Spanish ‘atención’ and ‘definir’); the English translations were produced by IRW, because the English printed by the study is the original English TMMS rather than a translation of the Spanish items administered.
zeng_2025_megaproject_ecm	Item codes follow the data file’s numbering, not the questionnaire’s. item_4_8 to item_4_15 are the paper’s eight-item SECM scale (questionnaire items 4.1-4.8). item_4_1 to item_4_7 are pollution-type items that appear in the data file but not in the published questionnaire, so their response labels are inferred from the shared rating grid, and any lead-in prompt they had is unknown.
zeng_2026_gai_exploitation	Administered in Chinese; the item text shown is the authors’ own English rendering published in the paper, as the Chinese wording was not released.
zeng_2026_gai_inst_support	Administered in Chinese; the item text shown is the authors’ own English rendering published in the paper, as the Chinese wording was not released.
zeng_2026_gai_learning_effect	Administered in Chinese; the item text shown is the authors’ own English rendering published in the paper, as the Chinese wording was not released.
zeng_2026_gai_role_adapt	Administered in Chinese; the item text shown is the authors’ own English rendering published in the paper, as the Chinese wording was not released.
zeng_2026_gai_self_efficacy	Administered in Chinese; the item text shown is the authors’ own English rendering published in the paper, as the Chinese wording was not released.
zeng_2026_gai_ttf	Administered in Chinese; the item text shown is the authors’ own English rendering published in the paper, as the Chinese wording was not released.
zhang_2020_trait_creativity_mood	Item tired.1 is labelled ‘tired’ in the source data file (a duplicated column header); the paper lists that mood item as ‘sleepy’, and it is assigned ‘sleepy’ here because it sits sixth in the file, where the paper lists ‘sleepy’, and the file’s first five mood columns follow the paper’s item order (the last four do not) – the two ‘tired’ columns correlate 0.74 and the data cannot confirm which is which. Wording is the paper’s English rendering; the questionnaire was presumably administered in Chinese.
zhang_2024_attractiveness	Item wording is the authors’ English version of a questionnaire administered in Chinese, and each wording is matched to its item code only by its number in the paper’s appendix; the data cannot confirm that match.
zhang_2024_expertise	The Expertise items were administered in Chinese; this table carries the English wording from the study’s own appendix, and the order-based tie of the three near-synonymous items to EXP1-EXP3 could not be checked against the data.
zhang_2024_ftcd	As for zhang_2024_smoking_rationalisation: the English labels in the supplementary SPSS file, not the Chinese wording respondents read.
zhang_2024_gift_intention	Administered in Chinese; the item text shown is the authors’ own English rendering from the paper’s appendix, as the Chinese wording was not released.
zhang_2024_parasocial	Administered in Chinese; the item text shown is the authors’ own English rendering published in the paper’s S1 Appendix, as the Chinese wording was not released.
zhang_2024_smoking_rationalisation	The item text and response options come from the SPSS variable and value labels in the article’s supplementary data file, which are the authors’ English rendering of a questionnaire administered in Chinese to adult male smokers in Guiyang. No Chinese wording is published in the deposit or the article, so this is not the text respondents read. The mapping from item code to text is the file’s own and is exact; only the language is a substitute.
zhang_2024_streamer_dsp	Administered in Chinese; the item text is the authors’ English from the paper’s appendix, and because the paper says this scale was reverse-scored before analysis (the stored values match the paper’s post-reversal statistics) while the Chinese wording was not released, it cannot be determined whether resp 5 means agreement or disagreement with the English statements shown, so no response labels are given.
zhang_2024_viewer_dsp	Administered in Chinese; the item text shown is the authors’ own English rendering published in the paper’s S1 Appendix, as the Chinese wording was not released, and each wording is matched to its item code only by its number in that appendix.
zhang_2025_coercive_pressure	Administered in Chinese; the item text shown is the authors’ own English rendering from the paper’s appendix, as the Chinese wording was not released.
zhang_2025_environ_awareness	Administered in Chinese; the item text shown is the authors’ own English rendering from the paper’s appendix, as the Chinese wording was not released.
zhang_2025_green_supply_intent	The survey was administered in Chinese; the item text is the authors’ own English rendering from the paper’s appendix, as the Chinese wording was not published.
zhang_2025_normative_pressure	Administered in Chinese; the item text shown is the authors’ own English rendering from the paper’s appendix, as the Chinese wording was not released.
zhang_2025_self_efficacy	Administered in Chinese; the item text shown is the authors’ own English rendering from the paper’s appendix, as the Chinese wording was not released.
zhang_2026_ai_companion	The item text is the English wording of the deposit’s column headers. The survey was run in China and the administered language is inferred to be Chinese; the wording respondents read is not published.
zhao_2024_erq	The ERQ was administered in Chinese (the ERQ-C translation of Wang et al.), but neither the PLOS ONE paper nor its data supplement contains any item wording in either language – the data columns are bare codes ERQr1-ERQr10 – so the item text shown here is Gross and John’s (2003) canonical English ERQ rather than the sentences participants read. Item codes follow the instrument’s printed item numbers; the reappraisal (r) / suppression (s) split is confirmed by the data and every code reproduces the paper’s published factor loadings, but nothing in the data distinguishes items within a subscale by wording. Scale points 2, 3, 5 and 6 carry no option text because the ERQ form labels only 1, 4 and 7.
zhao_2025_digital_literacy	Administered in Chinese (CFPS); the item text is the authors’ English description from the paper, as the Chinese wording was not released. The paper’s Table 2 swaps the descriptions of digital_social and digital_entertainment relative to its own column names and Methods text; the item text here follows the column names (digital_social = socializing).
zhao_2025_leisure_involvement	The survey was administered in person at a botanical garden in Changsha, presumably in Chinese (the paper does not name the language); the item text is the authors’ English wording from the paper’s Table 1, the only wording published. The items are the authors’ adaptation of McIntyre and Pigram’s (1992) scale to the park setting, not the original wording.
zhao_2025_nat_env_perception	Administered in person at a botanical garden in Changsha, presumably in Chinese (the paper does not name the language); the item text shown is the authors’ English wording from the paper’s Table 1, the only wording published, and the paper gives no response-option labels.
zhao_2025_place_attachment	Administered in person at a botanical garden in Changsha, presumably in Chinese (the paper does not name the language); the item text shown is the authors’ English wording from the paper’s Table 1, the only wording published, and the paper gives no response-option labels.
zhao_2025_psych_recovery_eval	Administered in person at a botanical garden in Changsha, presumably in Chinese (the paper does not name the language); the item text shown is the authors’ English wording from the paper’s Table 1, the only wording published, and the paper gives no response-option labels.
zhou_2024_smart_home_intention	Administered to older adults in China, presumably in Chinese; the paper and its data file carry only the authors’ English wording, which is what is shown here. The construct coded ITS is intergenerational technical support (children’s help with the smart home), not intention to switch.
zhou_2025_ehealth_literacy	Administered to university students in China, presumably in Chinese; the study’s data file carries only the authors’ English wording of the eHealth Literacy Scale, which is what is shown here, and the paper names only the two end anchors of the 5-point scale.
zhu_2026_llm_meteorology_performance	Graduate students at a Chinese university rated LLM tools on five aspects in a survey run on a Chinese platform, but only an English rendering of the questions is published (the study’s own data-file headers), so IRW ships that English. Only the endpoints of the 1-5 scale are described (“1 is the lowest and 5 is the highest”), so response options carry no text.
znidarsic_2021_coworker_support	Respondents (employees in Slovenia) most likely answered a Slovenian-language questionnaire; the item text shown is the study’s own English from its data file, and the paper gives the top anchor as ‘very often’ in its Methods but ‘always’ in its Results.
znidarsic_2021_leader_support	Respondents (employees in Slovenia) most likely answered a Slovenian-language questionnaire; the item text shown is the study’s own English from its data file, and the paper gives the top anchor as ‘very often’ in its Methods but ‘always’ in its Results.
znidarsic_2021_org_practices	Respondents (employees in Slovenia) most likely answered a Slovenian-language questionnaire; the item text shown is the study’s own English from its data file, and the paper labels the endpoints ‘completely disagree’/‘strongly agree’ in its Methods but ‘strongly disagree’/‘very agree’ in its Results.
znidarsic_2021_wf_balance	Respondents (employees in Slovenia) most likely answered a Slovenian-language questionnaire, and the item text shown is the study’s own English from its data file; item WFB2 (I have problems with balancing work and non-work activities) is stored reverse-scored, so its anchors are shown reversed (1 = completely agree, 5 = completely disagree).